The ‘white-collar bloodbath’ is all part of the AI hype machine

https://www.cnn.com/2025/05/30/business/anthropic-amodei-ai-jobs-nightcap

by lwo32k • 1 year ago

713 1164 1 year ago

I look forward to the day where executive overpromises and engineering underdeliveries bring about another AI winter so the useful techniques can continue without the stench of the "AI" association and so the grifters go bankrupt.

sevensor3• 1 year ago

The implosion of this AI bubble is going to have a stupendous blast radius. It’s never been harder to distinguish AI from “things people do with computers” more generally. The whole industry is implicated, complicit, and likely to suffer when AI winter arrives. Dotcom bust didn’t just hit people who were working for pets.com.

pixl972• 1 year ago

Just like the internet was a fad, right?

DrillShopper6• 1 year ago

More like the dot-com bubble

threeseed4• 1 year ago

Internet only became a fad once it was already large and had tens of millions of users.

I remember the pre-Web days of Usenet and BBS and no one thought those were trendy.

AI is far more akin to crypto.

monero-xmr• 1 year ago

https://en.m.wikipedia.org/wiki/List_of_predictions_for_auto...

It wasn’t just Elon. The hype train on self driving cars was extreme only a few years ago, pre-LLM. Self driving cars exist sort of, in a few cities. Quibble all you want but it appears to me that “uber driver” is still a popular widespread job, let alone truck driver, bus driver, and “car owner” itself.

I really wish the AI ceos would actually make my life useful. For example, why am I still doing the dishes, laundry, cleaning my house, paying for landscaping, painters, and on and on? In terms of white collar work I’m paying my fucking lawyers more than ever. Why don’t they solve an actual problem

DrillShopper6• 1 year ago

In terms of white collar work I’m paying my fucking lawyers more than ever. Why don’t they solve an actual problem

Rule 0 is that you never put your angel investors out of work if you want to keep riding on the gravy train

Philpax• 1 year ago

Because textual data is plentiful and easy to model, and physical data is not. This will change - there are now several companies working on humanoid robots and the models to power them - but it is a fundamentally different set of problems with different constraints.

MangoToupe• 1 year ago

I really wish the AI ceos would actually make my life useful.

TBH, I do think that AI can deliver on the hype of making tools with genuinely novel functionality. I can think of a dozen ideas off the top of my head just for the most-used apps on my phone (photos, music, messages, email, browsing). It's just going to take a few years to identify how to best integrate them into products without just chucking a text prompt at people and generating stuff.

GardenLetter27• 1 year ago

Bureaucracy and regulation is the main issue there though.

Like in Europe where you're forced to pay a notary to start a business - it's not really even necessary, nevermind something that couldn't be automated, but it's just but of the establishment propping up bureaucrats.

Whereas LLMs and generative models in art and coding for example, help to avoid loads of bureaucracy in having to sort out contracts, or even hire someone full-time with payroll, etc.

jellicle• 1 year ago

We are going to have an ever-increasing supply of stories along the lines of "used a LLM to write a contract; contract gave away the company to the counterparty; now trying to get a court to dissolve the contract".

Sure you'll have destroyed the company, but at least you'll have avoided bureaucracy.

dosinga• 1 year ago

Like in Europe

Like in the US you have a choice of which jurisdiction you want to start your company. Not all require a notary

xxs• 1 year ago

Like in Europe where you're forced to pay a notary to start a business

Do you have a specific country in mind, as the statement is not true for quite a lot of EU member states... and likely untrue for most of the European countries.

edent• 1 year ago

Buy a dishwasher - they're cheap, work really well, and don't use much energy / water.

Same as a washing machine / drier. Chuck the clothes in, press a button, done.

There are Roomba style lawnmowers for your grass cutting.

I'll grant you painting a house and plumbing a toilet aren't there yet!

al_borland3• 1 year ago

With the laundry machine and dishwasher, it still requires effort. A human needs to collect the dirty stuff, put it into the machine properly, decide when it should run, load the soap, select a cycle type, start it, monitor the machine to know when it’s done, empty the machine, and put the stuff away properly, thus starting the human side of the process again.

It’s less work than it used to be, but remove the human who does all that and the dirty dishes and clothes will still pile up. It’s not like we have Rosie, from The Jetsons, handling all those things (yet). How long before the average person has robot servants at home? Until that day, we are effectively project managers for all the machines in our homes.

coffeefirst• 1 year ago

You know what I want? A LM that navigates customer support phone trees for me.

If you want to waste my time with an automated nonsense we should at least even the playing field.

This is feasible with today’s technology.

darthwalsh• 1 year ago

Sounds like Google Duplex, but I guess they never expanded the tech beyond restaurant reservations.

But on my Pixel now, on some phone trees it shows a UI with numbers and choices, and even predicts ahead for the other choices so you aren't forced to wait. Very handy!

Hilift2• 1 year ago

Self-driving cars are required to beep when in reverse. In both San Francisco and San Diego homes near Waymo charging facilities are a nuisance. The neighbors hate the beeping, and they operate late hours, and use things like shop vac cleaners that are loud. Whoever thought of this hates self driving cars and people. There is no way this can work in mixed urban areas.

darth_avocado6• 1 year ago

I don’t understand how any business leader can be excited about humans being replaced by AI. If no one has a job, who’s going to buy your stuff? When the unemployment in the country goes up, consumer spending slows down and recession kicks in. How could you be excited for that?

FeteCommuniste3• 1 year ago

I guess the idea is that the people left working will be made so productive and wealthy thanks to the miracle of AI that they can more than make up the difference with extravagant consumption.

isoprophlex• 1 year ago

I too plan to buy 100.000 liters of yogurt each day once AI has transported me into the socioeconomic strata of the 0.1%

darth_avocado6• 1 year ago

If you want to see what that looks like, just look at the economy of India. Do we really want that?

al_borland3• 1 year ago

A single rich person can only much door dash. Scaling a customer base needs to be done horizontally.

keybored2• 1 year ago

We have consumer capitalism now. Before we didn’t. There’s no reason it can’t be replaced.

Sure there can be rich people who are radical enough to push for another phase of capitalism.

That’s a kind of a capitalism which is worse for workers and consumers. With even more power in the hands of capitalists.

thmsths• 1 year ago

Tragedy of the commons: no one being able to buy stuff is a problem for everyone, but being able to save just a bit more by getting rid of your workforce is a huge advantage for your business.

JKCalhoun9• 1 year ago

… in the interim, of course.

bckr2• 1 year ago

“tragedy of the commons” is treated as a Theory of Human Nature when it’s really a religious principle underlying how we operate our society.

johnbenoe4• 1 year ago

You ever thought there’s more to life than work lol. Maybe humans can approach a new standard of living…

codr77• 1 year ago

Excellent choice of words there: new standard.

I'm sure we are, but it doesn't look like an improvement for most people.

darth_avocado6• 1 year ago

I’m yet to be convinced that if majority of the humans are out of work, the government will be able to take care of them and allow them to “pursue their calling”. Hunger games is a more believable outcome to me.

rfrey• 1 year ago

The most powerful nation on earth isn't even willing to extend basic health care to the masses, nevermind freeing them to pursue a higher calling than enriching billionaires.

JKCalhoun9• 1 year ago

If someone is going to suggest UBI, I wish they could explain to me how Reservations have failed so hard in the U.S.. I think that would be a cautionary tale.

onlyrealcuzzo2• 1 year ago

If no one has a job, who’s going to buy your stuff?

All the people employed by the government and blue collar workers? All the entrepreneurs, gig workers, black market workers, etc?

It's easy to imagine a world in which there are way less white collar workers and everything else is pretty much the same.

It's also easy to imagine a world in which you sell less stuff but your margins increase, and overall you're better off, even if everybody else has less widgets.

It's also easy to imagine a world in which you're able to cut more workers than everyone else, and on aggregate, barely anyone is impacted, but your margins go up.

There's tons of other scenarios, including the most cited one - that technology thus far has always led to more jobs, not less.

They're probably believing any combination of these concepts.

It's not guaranteed that if there's 5% less white-collar workers per year for a few decades that we're all going to starve to death.

In the future, if trends continue, there's going to be way less workers - since there's going to be a huge portion of the population that's old and retired.

You can lose x% of the work force every year and keep unemployment stable...

A large portion of the population wants a lot more people to be able to not work and get entitlements...

It's pretty easy to see how a lot of people can think this could lead to something good, even if you think all those things are bad.

Two people can see the same painting in a museum, one finds it beautiful, and the other finds it completely uninteresting.

It's almost like asking - how can someone want the Red team to win when I want the Blue team to win?

munksbeer5• 1 year ago

It's also easy to imagine a world in which you sell less stuff but your margins increase, and overall you're better off, even if everybody else has less widgets.

History seems to show this doesn't happen. The trend is not linear, but the trend is that we live better lives each century than the previous century, as our technology increases.

Maybe it will be different this time though.

darth_avocado6• 1 year ago

All the people employed by the government and blue collar workers

If people don’t have jobs, government doesn’t have taxes to employ other people. If CEOs are salivating at the thought of replacing white collar workers, there is no reason to think next step of AI augmented with robotics won’t replace blue collar workers as well.

JKCalhoun9• 1 year ago

All the people employed by the government and blue collar workers?

You forgot the born-wealthy.

I feel increasingly like a rube for having not made my little entrepreneurial side-gigs focused strictly on the ultra-wealthy. I used to sell tube amplifier kits, for example, so you and I could have a really high-end audio experience with a very modest outlay of cash (maybe $300). Instead I should have sold the same amps but completed for $10K. (There is no upper bounds for audio equipment though — I guess we all know.)

neutronicus2• 1 year ago

There are also blue- and pink-collar industries that we all tacitly agree are crazy understaffed right now because of brutal work conditions and low pay (health care, child care, K-12, elder care), with low quality-of-service a concern across the board, and with many job functions that seem very difficult to replace with AI (assuming liability for preventing children and elderly adults from physically injuring themselves and others).

If you, a CEO, eliminate a bunch of white-collar workers, presumably you drive your former employees into all these jobs they weren't willing to do before, and hey, you make more profits, your kids and aging parents are better-taken-care-of.

Seems like winning in the fundamental game of society - maneuvering everyone else into being your domestic servants.

spamizbad2• 1 year ago

All the people employed by the government and blue collar workers? All the entrepreneurs, gig workers, black market workers, etc?

I can tell you for many of those professions their customers are the same white collar workers. The blue collar economy isn't plumbers simply fixing the toilets of the HVAC guy, while the HVAC guy cools the home of the electrician, while...

ben_w3• 1 year ago

Game theory/Nash equilibrium/Prisoner's Dilemma, and the turkey's perspective in the problem of induction.

So far, for any given automation, each actor gets to cut their own costs to their benefit — and if they do this smarter than anyone else, they win the market for a bit.

Every day the turkey lives, they get a bit more evidence the farmer is an endless source of free food that only wants the best for them.

It's easy to fool oneself that the economics are eternal with reference to e.g. Jevons paradox.

JKCalhoun9• 1 year ago

turkey's perspective in the problem of induction…

Had to look that up: https://en.wikipedia.org/wiki/Turkey_illusion

abracadaniel• 1 year ago

My long term fear with AI is that by replacing entry level jobs, it breaks the path to train senior level employees. It could take a couple of decades to really feel the heat from it, but could lead to massive collapse as no one is left with any understanding of how existing systems work, or how to design replacements.

spacemadness2• 1 year ago

And we as humans figured all this out and still do nothing with this knowledge. We fight as hard as we can against collective wisdom.

absurdo3• 1 year ago

Basically if anyone has an iota of sensibility you should have never taken sama, Zuckerberg, Gates, or anyone else of that sort at face value. When they tell you they’re doing things for the good of humanity, look at what the other hand is up to.

roenxi4• 1 year ago

You're being confused by the numbers. We aren't trying to maximise consumer spending, the point is to maximise living standards. If the market equilibrium price of all goods was $0 consumer spending would be $0 and living standards would be off the charts. It'd be a great outcome.

It just happens that up to this point there have been things that couldn't be done by capital. Now we're entering a world where there isn't such a thing and it is unclear what that implies for the job market. But people not having jobs is hardly a bad thing as long as it isn't forced by stupid policy, ideally nobody has to work.

amanaplanacanal• 1 year ago

In theory. In reality, how are the benefits of all this efficiency going to be distributed to the people who aren't working? I sure don't see any calls for higher taxes and more wealth redistribution.

alluro22• 1 year ago

Given the current mechanics evident in the society - declining education, healthcare and rising cost of living, homelessness and exploding economic inequality - who is "we", trying to maximise living standards, and what movement do you see leading towards such an outcome?

untrust• 1 year ago

Another question: If AI is going to eat up everyone's jobs, how will any business be safe from a new competitor showing up and unseating them off their throne? I don't think that the low level peons would be the only ones at stake as a company could be easily outcompeted as well since AI could conceivably outperform or replace any existing product anyways.

I guess funding for processing power and physical machinery to run the AI backing a product would be the biggest barrier to entry?

layer82• 1 year ago

Institutional knowledge is key here. Third parties can’t replicate it quickly just by using AI.

zhobbs• 1 year ago

Yeah this will likely lead to margin compression. The best companies will be fine though, as brand and existing distribution is a huge moat.

JKCalhoun9• 1 year ago

The beginning of the AI Wars?

anvandare• 1 year ago

A cancerous cell does not care that it is (indirectly) killing the lifeform that it is a part of. It just does what it does without a thought.

And if it could think, it would probably be very proud of the quarter (hour) figures that it could present. The Number has gone up, time for a reward.

SpicyLemonZest10• 1 year ago

Business leaders in AI are _not_ excited and agree with your concerns. That's what the source article is about - the CEO of AI lab Anthropic said he sees major social problems coming soon. The problem is that the information environment is twisted in knots. The author, like many commentators, characterizes your concerns as "optimism" and "hype", because she doesn't think AI will actually have these large impacts.

spacemadness2• 1 year ago

I think he says this just to hype up how powerful of a force AI is which helps these CEOs bottom line eventually. Cynically “we’ve created something so powerful it will eliminate jobs and cause strife” gets those investors excited for more.

geraneum9• 1 year ago

They are. The audience of this talk is not normal people. He’s excited and is targeting a specific group in his messaging. The author is a person like majority.

leeroihe3• 1 year ago

They want an omnipresent, lobotomized and defeated underclass who only exists to "respond" to the ai to continue to improve it. This is basically what alexander wang from Scale AI explained at a recent talk which was frankly terrifying.

Your UBI will be controlled by the government, you will have even less agency than you currently have and a hyper elite will control the thinking machines. But don't worry, the elite and the government are looking out for your best interest!

pdfernhout• 1 year ago

We already have that "defeated underclass" courtesy of a century of mainstream schooling (according to NYS Teacher of the Year John Taylor Gatto): "The Underground History of American Education -- A conspiracy against ourselves" https://www.lewrockwell.com/2010/10/john-taylor-gatto/the-cu... "As soon as you break free of the orbit of received wisdom you have little trouble figuring out why, in the nature of things, government schools and those private schools which imitate the government model have to make most children dumb, allowing only a few to escape the trap. The problem stems from the structure of our economy and social organization. When you start with such pyramid-shaped givens and then ask yourself what kind of schooling they would require to maintain themselves, any mystery dissipates — these things are inhuman conspiracies all right, but not conspiracies of people against people, although circumstances make them appear so. School is a conflict pitting the needs of social machinery against the needs of the human spirit. It is a war of mechanism against flesh and blood, self-maintaining social mechanisms that only require human architects to get launched. I’ll bring this down to earth. Try to see that an intricately subordinated industrial/commercial system has only limited use for hundreds of millions of self-reliant, resourceful readers and critical thinkers. In an egalitarian, entrepreneurially based economy of confederated families like the one the Amish have or the Mondragon folk in the Basque region of Spain, any number of self-reliant people can be accommodated usefully, but not in a concentrated command-type economy like our own. Where on earth would they fit? In a great fanfare of moral fervor some years back, the Ford Motor Company opened the world’s most productive auto engine plant in Chihuahua, Mexico. It insisted on hiring employees with 50 percent more school training than the Mexican norm of six years, but as time passed Ford removed its requirements and began to hire school dropouts, training them quite well in four to twelve weeks. The hype that education is essential to robot-like work was quietly abandoned. Our economy has no adequate outlet of expression for its artists, dancers, poets, painters, farmers, filmmakers, wildcat business people, handcraft workers, whiskey makers, intellectuals, or a thousand other useful human enterprises — no outlet except corporate work or fringe slots on the periphery of things. Unless you do "creative" work the company way, you run afoul of a host of laws and regulations put on the books to control the dangerous products of imagination which can never be safely tolerated by a centralized command system...."

In 2010, I put together a list of alternatives here to address the rise of AI and Robotics and its effect on jobs: https://pdfernhout.net/beyond-a-jobless-recovery-knol.html "This article explores the issue of a "Jobless Recovery" mainly from a heterodox economic perspective. It emphasizes the implications of ideas by Marshall Brain and others that improvements in robotics, automation, design, and voluntary social networks are fundamentally changing the structure of the economic landscape. It outlines towards the end four major alternatives to mainstream economic practice (a basic income, a gift economy, stronger local subsistence economies, and resource-based planning). These alternatives could be used in combination to address what, even as far back as 1964, has been described as a breaking "income-through-jobs link". This link between jobs and income is breaking because of the declining value of most paid human labor relative to capital investments in automation and better design. Or, as is now the case, the value of paid human labor like at some newspapers or universities is also declining relative to the output of voluntary social networks such as for digital content production (like represented by this document). It is suggested that we will need to fundamentally reevaluate our economic theories and practices to adjust to these new realities emerging from exponential trends in technology and society."

carlosjobim2• 1 year ago

That's a very pessimistic view. People can borrow money against their property, then later they can borrow money against their diploma and professional certificates (and nobody should be allowed to work without being certified, that's dangerous). Then later I think it's time for banks to start offering consumers the reproductive right of mortgaging their children, either born or unborn.

tim33316• 1 year ago

It's like all the farmers soil shoveling jobs were stolen by tractors. People moved on to more interesting things.

lowbloodsugar2• 1 year ago

In all previous such revolutions, humans were freed to do more productive work while the cost of goods came down. But that doesn’t mean the same is true this time. Now the revolution does not make physical tasks easier (like ploughing or spinning thread) but intellectual labor. This time, there are no jobs to go to, since those jobs are also done by AI.

bravesoul27• 1 year ago

How to move to post-capitalism is the question. We need something money-like to motivate but something different.

keybored2• 1 year ago

If the CEO of a soda company declared that soda-making technology is getting so good it’s going to ruin the global economy, you’d be forgiven for thinking that person is either lying or fully detached from reality.

Exactly. These people are growth-seekers first, domain experts second.

Yet I saw progressive[1] outlets reacting to this as a neutral reporting. So it apparently takes a “legacy media” outlet to wake people out of their AI stupor.

[1] American news outlets that lean social-democratic

sevensor3• 1 year ago

What AI is going to wipe out is white collar jobs where people sleepwalk through the working day and carelessly half ass every task. In 2025, we can get LLMs to do that for us. Unfortunately, the kind of executive who thinks AI is a legitimate replacement for actual work does not recognize the difference. I expect to see the more credulous CEOs dynamiting their companies as a result. Whether the rest of us can survive this remains to be seen. The CEOs will be fine, of course.

Der_Einzige2• 1 year ago

Consulting companies like the Big 4 where this happens most are bigger/stronger than ever (primarily due to AI related consulting). Try again.

sevensor3• 1 year ago

What makes you think productive work is what consulting companies are selling? They're there for laundering accountability. When you bring in consultants to roll out your corporate AI strategy, and it all falls apart in a few years, you can say, "we were following best practices, nobody could have anticipated X," where X is whatever failure mode ultimately tanks the AI strategy.

code_for_monkey2• 1 year ago

I think this is the kind of logic you wind up with when you start with the assumption that the Big 4 tell the truth about absolutely everything all the time

airstrike2• 1 year ago

Consulting companies don't sell productive advice. They sell management insurance.

johnbenoe4• 1 year ago

Yea

psadauskas• 1 year ago

AIs are great at generating bullshit, so if your job involves generating bullshit, you're probably on the chopping block.

I just wish that instead of getting more efficient at generating bullshit, we could just eliminate the bullshit.

potatoman22• 1 year ago

Some of the best applications of LLMs I've seen are for reducing bullshit. My goal for creating AI products is to let us act more like humans and less like oxen. I know it's idealistic, but I need to act with some goal.

TeMPOraL• 1 year ago

AIs are great at generating bullshit, so if your job involves generating bullshit, you're probably on the chopping block.

That covers majority of sales, advertising and marketing work. Unfortunately, replacing people with AI there will only make things worse for everyone.

cjs_ac3• 1 year ago

There's a connection to the return to office mandates here: the managers who don't see how anyone can work at home are the ones who've never done anything but yap in the office for a living, so they don't understand how sitting somewhere quiet and just thinking counts as work or delivers value for the company. It's a critical failure to appreciate that different people do different things for the business.

Jubijub• 1 year ago

That is a hugely simplistic take that tells me you never managed people out coordinated work across many people. I mean I a more productive individually at home too, so are probably all my folks in the team. But we don’t always work independently from each others, by which point having some days in common is a massive booster

const_cast5• 1 year ago

What AI is going to wipe out is white collar jobs where people sleepwalk through the working day and carelessly half ass every task.

The only reason this existed in the first place is because measuring performance is extremely difficult, and becomes more difficult the more complex a person's job is.

AI won't fix that. So even if you eliminate 50% of your employees, you won't be eliminating the bottom 50%. At worst, and probably what happens on average, your choices are about as good as random choice. So you end up with the same proportion of shitty workers as you had before. At worst worst, you actively select the poorest workers because you have some shitty metrics, which happens more often than we'd all like to think.

richardw• 1 year ago

CEO’s will be fine until their customers disappear. Are the AI’s going to click ads and buy iPhones?

einpoklum3• 1 year ago

I haven't worked in the US; and - have not yet worked in a company where such employees exist. Some are slower, some are fast or more efficient or productive - but they're all, everyone, under the pressure of too many tasks assigned to them, and it's always obvious that more personnel is needed but budget (supposedly) precludes it.

So, what you're describing is a mythical situation for me. But - US corporations are fabulously rich, or perhaps I should say highly-valued, and there are lots of investors to throw money at things I guess, so maybe that actually happens.

ryandrake6• 1 year ago

No, it's the same in the US, too. I don't know what these mythical companies are where people are saying 50% of the workforce does nothing, but I've never seen such a place. Everywhere I've ever worked had way more projects to get done than people available to do them. Everyone was working at capacity.

xg15• 1 year ago

What AI is going to wipe out is white collar jobs where people sleepwalk through the working day and carelessly half ass every task.

Note that AI wipes out the jobs, but not the tasks themselves. So if that's true, as a consumer, expect more sleepwalked, half-assed products, just created by AI.

habosa• 1 year ago

I think it’s actually going to save those people. They can vibe code themselves just enough output to survive where before they did next to nothing. In relative terms, they’ll get a much much higher productivity boost from AI than the already high-performing Staff engineer.

Management will be thrilled.

throw2342342342• 1 year ago

Nah - those people have the bandwidth/time to justify their value in my experience. They are also usually the people managing the productive.

Its the people that are constantly working, and too busy to be seen, producing output and keeping the lights on who don't have time for the "games" who AI is going for. Their jobs are easier to define since they are productive and do "something" - so its easy to market AI products for these use cases. After all these people are usually not the people in charge of the purse strings in most organisations for better or worse.

whynotminot8• 1 year ago

There’s a hype machine for sure.

But the last few paragraphs of the piece kind of give away the game — the author is an AI skeptic judging only the current products rather than taking in the scope of how far they’ve come in such a short time frame. I don’t have much use for this short sighted analysis. It’s just not very intelligent and shows a stubborn lack of imagination.

It reminds me of that quote “it is difficult to get a man to understand something, when his salary depends on his not understanding it.”

People like this have banked their futures on AI not working out.

codr77• 1 year ago

The opposite is more true imo.

It's the AI hype squad that are banking their future on AI magically turning into AGI; because, you know, it surprised us once.

whynotminot8• 1 year ago

Not really — even if AGI doesn’t work and these models don’t get any better, there’s still enormous value to be mined just from harnessing the existing state of the art.

Or these guys pivot and go back to building CRUD apps. They’re either at the front of something revolutionary… or not… and they’ll go back to other lucrative big tech jobs.

bgwalter3• 1 year ago

Using the Upton Sinclair quote in this context is a sign of not understanding the quote. The original quote means that you ignore gross injustices of your employer in order to stay employed.

It was never used in the sense of denigrating potential competitors in order to stay employed.

People like this have banked their futures on AI not working out.

If "AI" succeeds, which is unlikely, what is your recommendation to journalists? Should they learn how to code? Should they become prostitutes for the 1%?

Perhaps the only option would be to make arrangements with the Mafia like dock workers to protect their jobs. At least it works: Dock workers have self confidence and do not constantly talk about replacing themselves. /s

whynotminot8• 1 year ago

I think the quote makes perfect sense in this context, regardless of the prior application.

As to my recommendation to what they do — I dunno man. I’m a software engineer. I don’t know what I am going to do yet. But I’m sure as shit not burying my head in the sand.

johnwheeler• 1 year ago

I previously worked at a company called Recharge Payments, directly supporting the CTO, Mike—a genuinely great person, and someone learning to program. Mike would assign me small tasks, essentially making me his personal AI assistant. Now, I approach everything I do from his perspective. It’s clear that over time, he’ll increasingly rely on AI, asking employees less frequently. Eventually, it’ll become so efficient to turn to AI that he’ll rarely need to ask employees anything at all.

lexandstuff2• 1 year ago

I've never had a job like that. My job has always involved helping my company, not just figure out how to build something, but what to build. We typically collaborate on a few ideas and then go away, let them percolate in our brains, before coming back with some new ideas to try. The whole point of the Agile Manifesto is that we don't know what to build in the first place.

Sometimes my boss has asked me to do something that in the long run will cost the company dearly. Luckily for him, I am happy to push back, because I can understand what we're trying to achieve and help figure the best option for the company based on my experience, intuition and the data I have available.

There's so much more to working with a team than: "Here is a very specific task, please execute it exactly as the spec says". We want ideas, we want opinions, we want bursts of creative inspiration, we want pushback, we want people to share their experiences, their intuition, the vibe they get, etc.

We don't want AI agents that do exactly what we say; we want teams of people with different skill sets who understand the problem and can interpret task through the lens of their skill set and experience, because a single person doesn't have all the answers.

I think your ex-boss Mike will very soon find himself trapped in local minima of innovation, with only his own understanding of the world, and a sycophantic yes-man AI employee that will always do exactly as he says. The fact that AI mostly doesn't work is only part of the problem.

golol• 1 year ago

To be clear, Amodei didn’t cite any research or evidence for that 50% estimate.

I truly belive these types of paper don't deserve to be valued so much.

righthand• 1 year ago

Yes we live in a world where no “experts” are required to provide any evidence or truth, but media outlets will gladly publish every false word and idea. For the same reason these Ceos want to wipe their workforce for more money, not a functioning society.

airstrike2• 1 year ago

The attention economy is ruining society.

madaxe_again6• 1 year ago

And the journalist cited what research or evidence, precisely, in his rebuttal?

idkwhattocallme4• 1 year ago

I worked at two different $10B+ market cap companies during ZIRP. I recall in most meetings over half of the knowledge workers attending were superfluous. I mean, we hired someone on my team to attend cross functional meetings because our calendars were literally too full to attend. Why could we do that? Because the company was growing and hiring someone to attend meetings wasn't going to hurt the skyrocketing stock. Plus hiring someone gave my VP more headcount and therefore more clout. The market only valued company growth, not efficiency. But the market always capitulates to value (over time). When that happens all those overlay hires will get axed. Both companies have since laid off 10K+. AI was the scapegoat. But really, a lot of the knowledge worker jobs it "replaces" weren't providing real value anyway.

bachmeier4• 1 year ago

I mean, we hired someone on my team to attend cross functional meetings because our calendars were literally too full to attend.

Some managers read Dilbert and think it's intended as advice.

trhway2• 1 year ago

AI has been also consuming Dilbert as part of its training...

phendrenad22• 1 year ago

To the contrary - they were providing value to the VP who benefitted from inflated headcount. That's "real value", it's just a rogue agent is misaligned with the company's goals.

And AI cannot provide that kind of value. Will a VP in charge of 100 AI agents be respected as much as a VP in charge of 100 employees?

At the end of the day, we're all just monkeys throwing bones in the air in front of a monolith we constructed. But we're not going to stop throwing bones in the air!

BriggyDwiggs422• 1 year ago

We really oughta work on setting up systems that don’t waste time on things like this. Might be hard, but probably would be worth the effort.

idkwhattocallme4• 1 year ago

True! I golfed with the president of the division on a Friday (during work) and we got to the root of this. Companies would rather burn money on headcount (counted as R&D) than show profits and pay the govt taxes. When you have 70%+ margin on your software, you have money to burn. Dividends back to shareholders was not rewarded during ZIRP. On VP's being respected. I found at the companies I worked at VPs and their directs were like Nobles in a feudal kingdom constantly quibbling/battling for territory. There were alliances with others and full on takeouts at points. One VP described it as Game of Thrones. Not sure how this all changes when your kingdom is a bunch of AI agents that presumably anyone can operate.

PeterStuer3• 1 year ago

"Hiring someone gave my VP more headcount and therefore more clout"

Which is the sole reason automation will not make most people obsolete until the VP level themselves are automated.

dlivingston• 1 year ago

No, not if the metric by which VPs get clout changes.

Nasrudith4• 1 year ago

Automation is just one form of "face a sufficiently competitive marketplace such that the company can no longer tolerate the dead-weight loss of their egos".

icedchai3• 1 year ago

I've worked at smaller companies where half the people in the meetings were just there because they had nothing else to do. Lots of "I'm a fly on the wall" and "I'll be a note taker" types. Most of them contributed nothing.

xp844• 1 year ago

My friend's company (he was VP of Software & IT at a non-tech company) had a habit of meetings with no particular agenda and no decisions that needed making. Just meeting because it was on the calendar, discussing any random thing someone wanted to blab about. Not how my friend ran his team but that was how the rest did.

Then they had some disappointing results due to their bad decision-making elsewhere in the company, and they turned to my friend and said "Let's lay off some of your guys."

Nasrudith4• 1 year ago

The first mistake is thinking that contribution must be in the form of output instead of ingestion. Of course meetings aren't often the most efficient form of doing so. More being forced to listen (at least officially) so there isn't an excuse.

disambiguation2• 1 year ago

But really, a lot of the knowledge worker jobs it "replaces" weren't providing real value anyway.

I think quotes around "real value" would be appropriate as well. Consider all the great engineering it took to create Netflix, valued at $500b - which achieves what SFTP does for free.

jsnider3• 1 year ago

Netflix's value comes from being convenient and compatible with the copyright system in a way sharing videos P2P definitely isn't.

paulcole• 1 year ago

Just curious, did you put yourself in the superfluous category either time?

idkwhattocallme4• 1 year ago

Ultimately (and sadly) yes. While I never habitually or intentionally attended meetings to just look busy, I did work on something I knew had a long shot of creating value for the business. I worked on 0-1 products that if the company was more disciplined would not (or should not) have attempted. I left both on my own accord seeing the writing on the wall.

mlsu• 1 year ago

I suspect that these "AI layoffs" are really "interest rate" layoffs in disguise.

Software was truly truly insane for a bit there. Straight out of college, no-name CS degree, making $120, $150k (back when $120k really meant $120k)? The music had to stop on that one.

catigula11• 1 year ago

That really only happened in HCOL areas.

spamizbad2• 1 year ago

Yeah, my spiciest take is that Jr. Dev salaries really started getting silly during the 2nd half of the 2010s. It was ultimately supply (too little) and demand (too much) pushing them upward, but it was a huge signal we were in a bubble.

nyarlathotep_4• 1 year ago

The irony now is that 120k is basically minimum wage for major metros (and in most cases that excludes home ownership).

Of course, that growth in wages in this sector was a contributing factor to home/rental price increases as the "market" could bear higher prices.

hn_throwaway_997• 1 year ago

This is so true. We had a (admittedly derogatory) term we used during the rise in interest rates, "zero interest rate product managers". Don't get me wrong, I think great product managers are worth their weight in gold, but I encountered so many PMs during the ZIRP era who were essentially just Jira-updaters and meeting-schedulers. The vast majority of folks I see that were in tech that are having trouble getting hired now are in people who were in those "adjacent" roles - think agile coaches, TPMs, etc. (but I have a ton of sympathy for these folks - many of them worked hard for years and built their skills - but these roles were always somewhat "optional").

I'd also highlight that beyond over-hiring being responsible for the downturn in tech employment, I think offshoring is way more responsible for the reduction in tech than AI when it comes to US jobs. Video conferencing tech didn't get really good and ubiquitous (especially for folks working from home) until the late teens, and since then I've seen an explosion of offshore contractors. With so many folks working remotely anyway, what does it matter if your coworker is in the same city or a different continent, as long as there is at least some daily time overlap (which is also why I've seen a ton of offshoring to Latin America and Europe over places like India).

catigula11• 1 year ago

Off-shoring is pretty big right now but what shocks me is that when I walk around my company campus I see obscene amounts of people visibly and culturally from, mostly, India and China. The idea that literally massive amounts of this workforce couldn't possibly be filled by domestic grads is pretty hard to engage with. These are low level business and accounting analyst positions.

Both sides of the aisle retreated from domestic labor protection for their own different reasons so the US labor force got clobbered.

icedchai3• 1 year ago

I worked at a small company with more PMs than developers. It was incredible how much bull it created.

boogieknite• 1 year ago

first job out of college i was one of these pms. luckily i figured it out quickly and would spend maybe 2 hours a day working, 6 hours a day teaching myself to program. i cant believe that job existed and they gave it to me. one of my teammates was moved to HR and he was distraught over how he actually had work to do

adamtaylor_13• 1 year ago

I’m realizing that 100% of all product managers I have ever worked with were just ZIRP-PMs.

I have never once worked with a product manager who I could describe as “worth their weight in gold”.

Not saying they don’t exist, but they’re probably even rarer than you think.

aswegs8• 1 year ago

How are TPMs optional? In my experience they provide more value than PMs that don't understand technology.

cavisne• 1 year ago

My theory for these PM's is its basically a cheap way to take potential entrepreneurs off the market. Its hard to predict if a startup will succeed but one genre of success is having a Type A "fake it till you make it" non technical cofounder who can keep raising long enough to get product market fit.

These types all go to the same schools and do really well, interview the same, and value the prestige of working in big tech. So it's pretty easy to identify them and offer them a great career path and take them off the market.

Technical founders are way trickier to identify as they can be dropouts, interview poorly, not value the prestige etc.

lukev• 1 year ago

Whenever I think about AI and labor, I can't help thinking about David Graeber's [Bullshit Jobs](https://en.wikipedia.org/wiki/Bullshit_Jobs).

And there's multiple confounding factors at play.

Yes, lots of jobs are bullshit, so maybe AI is a plausible excuse to downside and gain efficiency.

But also the dynamic that causes the existence of bullshit jobs hasn't gone away. In fact, assuming AI does actually provide meaningful automation or productivity improvemenet, it might well be the case that the ratio of bullshit jobs increases.

alvah2• 1 year ago

Exactly. For as long as I can remember, in any organisation of any reasonable size I have worked in, you could get rid of the ~50% of the headcount who aren't doing anything productive without any noticeable adverse effects (on the business at least, obviously the effects on the individuals would be somewhat adverse). This being the case, there are obviously many other factors other than pure efficiency keeping people employed, so why would an AI revolution on it's own create some kind of massive Schumpeterian shockwave?

throw2342342342• 1 year ago

Its hard to automate something that is hard to define, so I see the productive jobs/workers being punished by AI moreso than those jobs. Generally I see anecdotally:

- Value creators (i.e. the ones historically carrying companies with the 80%/20% rule) generally are the ones cautious and/or fearful of AI. The ones that carried most of the company. Their output is measurable and definable so able to be automated.

- The people in the jobs you mention in your post conversely are usually the ones most excited about AI. The ones in meetings all day, in the corporate machine. By definition their job is already not well defined anyway - IMV this is harder to automate. They are often there for other reasons other than "productive output" - e.g. compliance, nepotism, stakeholder management, etc.

ozim• 1 year ago

Bad part is all those guys attending meetings start feeling important. They start feeling like they are doing the job.

I’ve seen those guys it is painful to watch.

JSR_FDED• 1 year ago

I don’t doubt there’s a lot of knowledge workers who aren’t adding value.

I’m worried about the shrinking number of opportunities for juniors.

hn_throwaway_997• 1 year ago

I agree with this, but I still think that offshoring is much more responsible for this than AI.

I have definitely seen real world examples where adding junior hires at ~$100k+ is being completely forgone when you can get equivalent output from someone making $40k offshore.

matthest• 1 year ago

Does anyone else think the fact that companies hire superfluous employees (i.e. bullshit jobs) is actually fantastic?

Because they don't have to do that. They could just operate at max efficiency all the time.

Instead, they spread the wealth a bit by having bullshit jobs, even if the existence of these jobs is dependent on the market cycle.

nyarlathotep_4• 1 year ago

Does anyone else think the fact that companies hire superfluous employees (i.e. bullshit jobs) is actually fantastic?

I do.

It's much more important that people live a dignified life and be able to feed their families than "increasing shareholder value" or whatever.

I'm a person that would be hypothetically supportive of something like DOGE cuts, but I'd rather have people earning a living even with Soviet-style make work jobs than unemployed. I don't desire to live in a cutthroat "competitive" society where only "talent" can live a dignified life. I don't know if that's "wealth distribution" or socialism or whatever; I don't really care, nor make claim it's some airtight political philosophy.

827a• 1 year ago

Half of everyone at most large companies could be retired with no significant impact to the company's ability to generate revenue. The problem has always been figuring out which half.

daxfohl2• 1 year ago

Agree, but two questions:

First, is AI really a better scapegoat? "Reducing headcount due to end of ZIRP" maybe doesn't sound great, but "replacing employees with AI" sounds a whole lot worse from a PR perspective (to me anyway).

Second, are companies actually using AI as the scapegoat? I haven't followed it too closely, but I could imagine that layoffs don't say anything about AI at all, and it's mostly media and FUD inventing the correlation.

ledauphin• 1 year ago

the one does actually sound worse because... it's actually worse. it clarifies that the companies themselves were playing games with people's livelihoods because of the potential for profit.

whereas "AI" is intuitively an external force; it's much harder to assign blame to company leadership.

leflambeur4• 1 year ago

isn't the scapegoat he or she who gets sacrificed? I think engineers are that

__turbobrew__2• 1 year ago

Turns out 50% of white collar jobs are just daycare for adults.

ivape7• 1 year ago

I've said this many times, that the abundance and wealth of the tech industry basically provided vast amounts of Universal Basic Income to a variety of roles (all of agile is one example). We're at a critical moment where we actually have to look at cost-cutting on this UBI.

federiconafria• 1 year ago

"my VP more headcount and therefore more clout"

This had me thinking, how are they going to get "clout", by comparing AI spending?

paulluuk2• 1 year ago

Around the time when bitcoin started to get serious public attention, late 2017, I remember feeling super hyped about it and yet everyone told me that money spent on bitcoin was wasted money. I really believed that bitcoin, or at least cryptocurrency as a whole, would fundamentally change how banking and currencies would work. Now, almost 10 years later, I would say that it did not live up to my believe that it would "fundamentally" change currencies and banking. It made some minor changes, sure, but if it weren't for the value of bitcoin, it would still be a nerdy topic about as well known as perlin noise. Although I did make quite a lot of money from it, though I sold out way too soon.

As a research engineer in the field of AI, I am again getting this feeling. People keep doubting that AI will have any kind of impact, and I'm absolutely certain that it will. A few years ago people said "AI art is terrible" and "LLMs are just autocomplete" or the famous "AI is just if-else". By now it should be pretty obvious to everyone in the tech community that AI, and LLMs in particular, are extremely useful and already have a huge impact on tech.

Is it going to fulfill all the promises made by billionaire tech CEOs? No, of course not, at least not on the time scale that they're projecting. But they are incredibly useful tools that can enhance efficiency of almost any job that involves setting behind a computer. Even just something like copilot autocomplete or talking with an LLM about a refactor you're planning, is often incredibly useful. And the amount of "intelligence" that you can get from a model that can actually run on your laptop is also getting much better very quickly.

The way I see it, either the AI hype will end up like cryptocurrency: forever a part of our world, but never quite lived up to it's promises, but I made a lot of money in the meantime. Or the AI hype will live up to it's promises, but likely over a much longer period of time, and we'll have to test whether we can live with that. Personally I'm all for a fully automated luxury communism model for government, but I don't see that happening in the "better dead than red" US. It might become reality in Europe though, who knows.

paulluuk2• 1 year ago

On a side note, I do worry about the energy consumption of AI. I'll admit that, like the silicon valley tech bros, there is a part of me that hopes that AI will allow researchers to invent a solution to that -- something like fusion or switching to quantum-computing AI models or whatever. But if that doesn't happen, it's probably the biggest problem related to AI. More so even than alignment, perhaps.

layer82• 1 year ago

already have a huge impact on tech

As a user, I haven’t seen a huge impact yet on the tech I use. I’m curious what the coming years will bring, though.

jollyllama• 1 year ago

Crypto is a really interesting point, because even the subset of people who have invested in it don't use it on a day to day basis. The entire valuation is based on speculative use cases.

surgical_fire• 1 year ago

Something can be useful and massively overhyped at the same time.

LLMs are good productivity tools. I've been using it for coding, and it is massively helpful, really speeds things up. There's a few asterisks there though

1) I does generate bullshit, and this is an unavoidable part of what LLMs are. The ratio of bullshit seems to come down with reasoning layers above it, but they will always be there.

2) LLMs, for obvious reasons, tend to be more useful the more mainstream languages and libraries I am working with. The more obscure it is, the less useful it gets. It may have a chilling effect on technological advancement - new improved things are less used because LLMs are bad at them due to lack of available material, the new things shrivel and die on the vine without having a chance of organic growth.

3) The economics of it are super unclear. With the massive hype there's a lot of money slushing around AI, but those models seem obscenely expensive to create and even to run. It is very unclear how things will be when the appetite of losing money at this wanes.

All that said, AI is multiple breakthroughs away of replacing humans, which does not mean they are not useful assistants. And increase in productivity can lead to lower demand for labor, which leads ro higher unemployment. Even modest unemployment rates can have grim societal effects.

The world is always ending anyway.

rvz• 1 year ago

By now it should be pretty obvious to everyone in the tech community that AI, and LLMs in particular, are extremely useful and already have a huge impact on tech.

Enough to cause the next financial crash, achieving a steady increase of 10% global unemployment in the next decade at worst,

That is the true definition of AGI.

kayamon• 1 year ago

Bitcoin was $2000 dollars each in 2017. Now in 2025 it's $104,000. It's set to keep countering global inflation until 2140.

It ain't done yet.

fny4• 1 year ago

I think everyone is missing the bigger picture.

This is not a matter of whether AI will replace humans whole sale. There are two more predominant effects:

1. You’ll need fewer humans to do the same task. In other forms of automation, this has led to a decrease in employment. 2. The supply of capable humans increases dramatically. 3. Expertise is no longer a perfect moat.

I’ve seen 2. My sister nearly flunked a coding class in college, but now she’s writing small apps for her IT company.

And for all of you who poo poo that as unsustainable. I became proficient in Rust in a week, and I picked up Svelte in a day. I’ve written a few shaders too! The code I’ve written is pristine. All those conversations about “should I learn X to be employed” are totally moot. Yes APL would be harder, but it’s definitely doable. This is an example of 3.

Overall, this will surely cause wage growth to slow and maybe decrease. In turn, job opportunities will dry up and unemployment might ensue.

For those who still don’t believe, air traffic controllers are a great thought experiment—they’re paid quite nicely. What happens if you build tools so that you can train and employ 30% of the population instead of just 10%?

BigJono• 1 year ago

I became proficient in Rust in a week, and I picked up Svelte in a day. I’ve written a few shaders too! The code I’ve written is pristine. All those conversations about “should I learn X to be employed” are totally moot.

fucking lmao

fny4• 1 year ago

My point is you learn X and your time to learn and ship Y is dramatically reduced.

It would have taken me a month to write the GPU code I needed in Blender, and I had everything working in a week.

And none of this was "vibed": I understand exactly what each line does.

whyowhy34849393• 1 year ago

Yes, this is definitely missing a /s, I hope.

Please for the love of god tell me this is a joke.

hooverd3• 1 year ago

Can you talk about Rust without your friend computer?

fny4• 1 year ago

Of course not! But I can definitely ship useful tools, and I can could learn to talk the talk in a tenth of the time it would otherwise have taken.

Which is my point, this is not about replacement, it's about reducing the need and increasing supply.

MattSayar2• 1 year ago

Can you talk about assembly without the internet?

I fully understand your point and even agree with it to an extent. LLMs are just another layer of abstraction, like C is an abstraction for asm is an abstraction for binary is an abstraction for transistors... we all stand on the shoulders of giants. We write code to accomplish a task, not the other way around.

ironman1478• 1 year ago

"I became proficient in Rust in a week". How did you evaluate that if you weren't an expert in Rust to begin with? What does proficient mean to you? Also, are you advocating we get rid of air traffic controllers with AI? How would we train the AI? What model would you use? If you can't solve a safety critical problem from first principles, there is no way an AI should be in the loop. This makes no sense.

Cynically, I'm happy we have this AI generated code. It's gonna create so much garbage and they'll have to pay good senior engineers more money to clean it all up.

ofjcihen4• 1 year ago

To your second point we’re seeing a huge comeback of vulnerabilities that we’re “mostly gone”. Things like very basic RCEs and SQLi. This is a great thing for security workers as well.

stefan_• 1 year ago

I don't understand, no one ever needed an LLM to automate air traffic controllers. 1980s tech could do that just fine. The reason they continue to exist is essentially cultural. Fell into a local maximum trap and now the entire industry and governance is incapable of lifting itself out of it and instead come up with stuff like "standardized phrases for the voice coms that we have inexplicably made crucial to the entire system" while riding cultural cliches like "the pilot must be in control" as they continue manual flight into big rocks.

lexandstuff2• 1 year ago

Re the last sentence, is the answer that more people will die in aviation disasters?

AnimalMuppet2• 1 year ago

At least temporarily, it can be somewhat self-fulfilling, though. Companies believe it, think they'd better shed white-collar jobs to stay competitive. If enough companies believe that, white-collar jobs go down, even if AI is useless.

Of course, in the medium term, those companies may find out that they needed those people, and have to hire, and then have to re-train the new people, and suffer all the disruption that causes, and the companies that didn't do that will be ahead of the game. (Or, they find out that they really didn't need all those people, even if AI is useless, and the companies that didn't get rid of them are stuck with a higher expense structure. We'll see.)

josefritzishere• 1 year ago

I don't think we've seen a technology more over-hyped in the history of industrialized society. Cars, which did fully replace horses, was not even hyped this hard.

spcebar• 1 year ago

Something is nagging me about the AI-human replacement conversation that I would love insight from people who know more about startup money than me. It seems like the AI revolution hit as interest rates went insane, and at the same time the AI that could write code was becoming available, the free VC money dried up, or at least changed. I feel like that's not usually a part of the conversation and I'm wondering if we would be having the same conversation if money for startups was thrown around (and more jobs were being created for SWEs) the way it was when interest rates were zero. I know next to nothing about this and would love to hear informed opinions.

swyx• 1 year ago

its not part of the conversation because the influence here is tangential at best (1) and your sense of how much vc money is on the table at any given time is not good (2).

1a. most seed/A stage investing is acyclical because it is not really about timing for exits, people just always need dry powder

1b. tech advancement is definitely acyclical - alexnet, transformers, and gpt were all just done by very small teams without a lot of funding. gpt2->3 was funded by microsoft, not vc

2a. (i have advance knowledge of this bc i've previewed the keynote slides for ai.engineer) free vc money slowed in 2022-2023 but has not at all dried up and in fact reaccelerated in a very dramatic way. up 70% this yr

2b. "vc" is a tenous term when all biglabs are >>10b valuation and raising from softbank or sovereign wealth. its no longer vc, its about reallocating capital from publics to privates because the only good ai co's are private

mjburgess5• 1 year ago

I'm not seeing how you're replying to this comment. I'm not sure you've understood their point.

The point is that there's a correlation between macroeconomic dynamics (ie., the price of credit increasing) and the "rise of AI". In ordinary times, absent AI, the macroeconomic dynamics would fully explain the economic shifts we're seeing.

So the question is why do we event need to mention AI in our explanation of recent economic shifts?

What phenomena, exactly, require positing AI disruption?

sfRattan• 1 year ago

It seems like the AI revolution hit as interest rates went insane...
...I'm wondering if we would be having the same conversation if money for startups was thrown around (and more jobs were being created for SWEs) the way it was when interest rates were zero.

The end of free money probably has to do with why C-level types are salivating at AI tools as a cheaper potential replacement for some employees, but describing the interest rates returning to nonzero percentages as going insane is really kind of a... wild take?

The period of interest rates at or near zero was a historical anomaly [1]. And that policy clearly resulted in massive, systemic misallocation of investment at global scale.

You're describing it as if that was the "normal?"

[1] https://www.macrotrends.net/2015/fed-funds-rate-historical-c...

michaeldoron2• 1 year ago

Every time an analyst gives the current state of AI-based tools as evidence supporting AI disruption being just a hype, I think of skeptics who dismissed the exponential growth of covid19 cases due to their initial low numbers.

Putting that aside, how is this article called an analysis and not an opinion piece? The only analysis done here is asking a labor economist what conditions would allow this claim to hold, and giving an alternative, already circulated theory that AI companies CEOs are creating a false hype. The author even uses everyday language like "Yeaaahhh. So, this is kind of Anthropic’s whole ~thing.~ ".

Is this really the level of analysis CNN has to offer on this topic?

They could have sketched the growth in foundation model capabilities vs. finite resources such as data, compute and hardware. They could have wrote about the current VC market and the need for companies to show results and not promises. They could have even wrote about the giant biotech industry, and its struggle with incorporating novel exciting drug discovery tools with slow moving FDA approvals. None of this was done here.

bckr2• 1 year ago

That’s not what major news outlets are for. I’m not sure exactly what they’re for.

SoftTalker4• 1 year ago

Analysis == Opinion when it comes to mainstream news reporting. It's one guy's thinking on something.

TheOtherHobbes• 1 year ago

I'm not seeing how comparing AI to a virus that killed millions and left tens of millions crippled is an effective way to support your argument.

drewcon2• 1 year ago

Humans are not familiar with exponential change so they have almost no ability to manage through exponential change.

Its an apt comparison. The criticisms in the cnn article are already out date in many instances.

dingnuts• 1 year ago

especially when the world population is billions and at the beginning we were worried about double digit IFR.

Yeah. Imagine if COVID had actually killed 10% of the world population. Killing millions sucks, but mosquitos regularly do that too, and so does tuberculosis, and we don't shut down everything. Could've been close to a billion. Or more. Could've been so much worse.

IshKebab• 1 year ago

I think you missed the point. AI is dismissed by idiots because they are looking at its state now, not what it will be in future. The same was true in the pandemic.

aaronbaugher• 1 year ago

Is this really the level of analysis CNN has to offer on this topic?

Not just this topic.

bgwalter3• 1 year ago

Why not use the promised exponential growth of home ownership that led to the catastrophic real estate bubble that burst in 2008 as an example?

We are still dealing with the aftereffects, which led to the elimination of any working class representation in politics and suppression of real protests like Occupy Wall Street.

When this bubble bursts, the IT industry will collapse for some years like in 2000.

michaeldoron2• 1 year ago

The growth of home ownership was an indicator of real estate investment, not of real world capabilities - once the value of real estate dropped and the bubble burst, those investments were worth less than before, causing the crisis. In contrast, the growth in this scenario is the capabilities of foundation models (and to a lesser extent, the technologies that stem out of these capabilities). This is not a promise or an investment, it's not an indication of speculative trust in this technology, it is a non-decreasing function indicating a real increase in performance.

PeterStuer3• 1 year ago

"Is this really the level of analysis CNN has to offer on this topic?"

It's not CNN exlusive. Newsmedia that did not evolve towards clicks, riling up people, hatewatching and paid propaganda to the highest bidder went extinct a decade ago. This is what did evolve.

biophysboy2• 1 year ago

This is outdated. Most of journalism has shifted to subscription models, offering a variety of products under one roof: articles, podcasts, newsletters, games, recipes, product reviews, etc.

biophysboy2• 1 year ago

Its an article reformulated from a daily newsletter. Newsletters take the form of a quick, casual follow up to current events (e.g. an Amodei interview). Its not intended to be exhaustive analysis.

Besides the labor economist bit, it also makes the correct point that tech people regularly exaggerate and lie. A great example of this is biotech, a field I work in.

leeroihe3• 1 year ago

The best heuristic is what people are realizing happened with uncheck "skilled" immigration in places like canada (and soon the U.S.). Everyone was sold that we "need these workers" because nobody was willing to work and that they added to GDP. When in reality, there's now significant evidence that all these new arrivals did was put a net drain on welfare, devalue the labor of endemic citizens (regardless of race - in many cases affecting endemic minorities MORE) and in the end, just reduced cost while degrading companies who did this.

We will wake up in 5 yrs to find we replaced people for a dependence on a handful of companies that serve llms and make inference chips. Its beyond dystopian.

matteotom• 1 year ago

Can you provide more details about said "significant evidence"? This seems to be a pretty popular belief, despite being contrary to generally accepted economics, and I've yet to see good evidence for it.

mjburgess5• 1 year ago

You can pick and choose problems from history where folk belief was wrong: WW1 vs. Y2K.

This isn't very informative. Indeed, engaging in this argument-by-analoguy betrays a lack of actual analysis, credible evidence and justification for a position. Arguing "by analogy" in this way, which picks and chooses an analogy, just restates your position -- it doesnt give anyone reasons to believe it.

timr• 1 year ago

I think of skeptics who dismissed the exponential growth of covid19 cases due to their initial low numbers.

Uh, not to be petty, but the growth was not exponential — neither in retrospect, nor given what was knowable at any point in time. About the most aggressive, correct thing you could’ve said at the time was “sigmoid growth”, but even that was basically wrong.

If that’s your example, it’s inadvertently an argument for the other side of the debate: people say lots of silly, unfounded things at Peak Hype that sound superficially correct and/or “smart”, but fail to survive a round of critical reasoning. I have no doubt we’ll look back on this period of time and find something similar.

Terr_2• 1 year ago

I think of skeptics who dismissed the exponential growth of covid19 cases due to their initial low numbers.

Compare: "Whenever I think of skeptics dismissing completely novel and unprecedented outcomes occurring by mechanisms we can't clearly identify or prove (will) exist... I think of skeptics who dismissed an outcome that had literally hundreds of well-studied historical precedents using proven processes."

You're right that humans don't have a good intuition for non-linear growth, but that common thread doesn't heal over those other differences.

actuallyalys2• 1 year ago

Yeah, for this analogy to work, we’d have to see AI causing a small but consistently doubling amount of lost jobs.

monkeyelite3• 1 year ago

I think of skeptics who dismissed the exponential growth of covid19 cases due to their initial low numbers.

But that didn’t happen. All of the people like pg who drew these accelerating graphs were wrong.

In fact, I think just about every commenter on COVID was wrong about what would happen in the early months regardless of political angle.

tim33316• 1 year ago

I remember scientists, especially epidemiologists being quite accurate. I guess the key is to not even have a political angle but instead some knowledge of what you are talking about.

qgin3• 1 year ago

This is the exact thing I’ve expressed as well.

This moment feels exactly to me like that moment when we were going to “shut down for two weeks” and the majority of people seemed to think that would be the end of it.

It was clear where the trend was going, but exponentials always seem ridiculous on an intuitive level.

deadbabe2• 1 year ago

It goes both ways. Once the exponential growth of COVID started, I heard wildly outrageous predictions of what was going to happen next, none of which ever really came to fruition.

bachmeier4• 1 year ago

AI is starting to get better than humans at almost all intellectual tasks

"Starting" is doing a hell of lot of work in that sentence. I'm starting to become a billionaire and Nobel Prize winner.

Anyway, I agree with Mark Cuban's statement in the article. The most likely scenario is that we become more productive as AI complements humans. Yesterday I made this comment on another HN story:

"Copilot told me it's there to do the "tedious and repetitive" parts so I can focus my energy on the "interesting" parts. That's great. They do the things every programmer hates having to do. I'm more productive in the best possible way.

But ask it to do too much and it'll return error-ridden garbage filled with hallucinations, or just never finish the task. The economic case for further gains has diminished greatly while the cost of those gains rises."

JKCalhoun9• 1 year ago

I'm starting to become a billionaire

Suggests you are accumulating money, not losing it. That I think is the point of the original comment: AI is getting better, not worse. (Or humans are getting worse? Ha ha, not ha ha.)

bachmeier4• 1 year ago

That I think is the point of the original comment: AI is getting better, not worse.

Well, in order to meet the standard of the quote "wipe out half of all entry-level office jobs … sometime soon. Maybe in the next couple of years" we need more than just getting better. We need considerably better technology with a better cost structure to wipe out that many jobs. Saying we're starting on that task when the odds are no better than me becoming a billionaire within two years is what we used to call BS.

SoftTalker4• 1 year ago

It it sustainable? I know when I program, it's sometimes nice to get to something that's easy, even if it's tedious and repetitive. It's like stopping to walk for a bit when you're on a run. You're still moving, but you can catch your breath and recharge.

bachmeier4• 1 year ago

Oh, I agree, but I'd say that it's probably easier to do those small things than it is to figure out a prompt to have Copilot do them. If it feels good, there's no reason not to do it yourself. I think we'd all agree that it's a joy to be able to tell Copilot to write out the scaffolding at the start of a new project.

ArtTimeInvestor• 1 year ago

Imagine you had a crystal ball that lets you look 10 years into the future, and you asked it about whether we underestimate or overestimate how many jobs AI will replace in the future.

It flickers for a moment, then it either says

"In 2025, mankind vastly underestimated the amount of jobs AI can do in 2035"

"In 2025, mankind vastly overestimated the amount of jobs AI can do in 2035"

How would you use that information to invest in the stock market?

elcapitan• 1 year ago

If I had a crystal ball that lets me look 10 years into the future and I wanted to invest in the stock market, I would ask it about the stock market.

JKCalhoun9• 1 year ago

I'm already assuming the first answer but nonetheless have absolutely no idea how I would use that to make a guess about the stock market.

So it's index funds (as always) with me anyway.

usersouzana• 1 year ago

Heads or tails, then proceed accordingly. You won't waste any more time analyzing it in hopes of getting it right.

dehrmann2• 1 year ago

Ah, so a straddle.

punkbit• 1 year ago

In that hypothetical case, you’d invest in the robot industry and security. The poor would try to eat the rich, right?

phendrenad22• 1 year ago

These are the moments that make millionaires. A majority of people believe that AI is going to thoroughly disrupt society. They've been primed to worry about an "AI apocalypse" by Hollywood for their entire lives. The prevailing counter-narrative is that AI is going to flop. HARD. You can't get more diametrically opposed than that. If you can correctly guess (or logically determine) which is correct, and bet all of your money on it, you can launch yourself into a whole other echelon of life.

I've been a heavy user of AI ever since ChatGPT was released for free. I've been tracking its progress relative to the work done by humans at large. I've concluded that it's improvements over the last few years are not across-the-board changes, but benefit specific areas more than others. And unfortunately for AI hype believers, it happens to be areas such as art, which provide a big flashy "look at this!" demonstration of AI's power to people. But... try letting AI come up with a nuanced character for a novel, or design an amplifier circuit, or pick stocks, or do your taxes.

I'm a bit worried about YCombinator. I like Hacker News. I'm a bit worried that YC has so much riding on AI startups. After machine learning, crypto, the post-Covid 19 healthcare bubble, fintech, NFTs, can they take another blow when the music stops?

SoftTalker4• 1 year ago

The prevailing counter-narrative is that AI is going to flop. HARD.

Why is that the counter-narrative? Doesn't it seem more likely that it will contine to gradually improve, perhaps asymptotically, maybe be more specifically trained in the niches where it works well, and it will just become another tool that humans use?

Maybe that's a flop compared to the hype?

ls6122• 1 year ago

At the rate the hyperscalers are increasing capex anything less than 1990s internet era growth rates will not be pretty. So far its been able to sustain those growth rates at the big boy AI companies (look at OpenAI revenue over time) but will it continue? Are we near the end of major LLM advances or are we near the beginning? There are compelling arguments both ways (running out of data is IMO the most compelling bear argument).

ryandrake6• 1 year ago

I wouldn't worry too much about YCombinator. Although individual investors can get richer or poorer, "investors" as a class effectively have unlimited money. Collectively, they will always be looking for a place to put it so it keeps growing even more, so there will always be work for firms like YCombinator to sprinkle all that investment money around.

tokioyoyo• 1 year ago

Not the biggest fan of crypto companies, but YC probably did well because of Coinbase.

barchar2• 1 year ago

It's not really enough to predict the outcome, you need something concrete to actually bet on, and you need to time things right (particularly for the pessimistic bet).

For any bet that involves purchasing bits of profits you you could be right and lose money because because the government generally won't allow the entire economy to implode.

By the time a bubble pops literally everyone knows they're in a bubble, knowing something is a bubble doesn't make it irrational to jump on the bandwagon.

ramesh312• 1 year ago

The prevailing counter-narrative is that AI is going to flop. HARD. You can't get more diametrically opposed than that.

The answer (as always) lies somewhere in the middle. Expert software developers who embrace the tech whole heartedly while understanding its' limitations are now in an absolute golden era of being able to do things they never could have dreamed of before. I have no doubt we will see the first unicorns made of "single pizza" size teams here shortly.

rjurney• 1 year ago

Workers in denial are like lemmings, headed for the cliff... not putting myself above that. A moderate view indicates great disruption before new jobs replace the current round being lost.

1vuio0pswjnm72• 1 year ago

"If the CEO of a soda company declared that soda-making technology is getting so good it's going to ruin the global economy, you'd be forgiven for thinking that person is either lying or fully detached from reality.

Yet when tech CEOs do the same thing, people tend to perk up."

Silicon Valley and Redmond make desperate attempts to argue for their own continued relevance.

For Silicon Valley VC, software running on computers cannot be just a tool. It has to cause "disruption". It has to be "eating the world". It has to be a source of "intelligence" that can replace people.

If software and computers are just boring appliances, like yesterday's typewriters, calculators, radios, TVs, etc., then Silicon Valley VC may need to find a new line of work. Expect the endless media hype to continue.

No doubt soda technology is very interesting. But people working at soda companies are not as self-absorbed, detached from reality and overfunded as people working for so-called "tech" companies.

digianarchist• 1 year ago

I saw a tweet the other day that stated AI will cure all diseases within 5-10 years. The tweet cites scientists and CEOs but only lists CEOs of AI companies.

https://x.com/kimmonismus/status/1927843826183589960

cadamsdotcom• 1 year ago

CEOs’ jobs involve hyping their companies. It’s up to us whether we believe.

I’d love a journalist using Claude to debunk Dario: “but don’t believe me, I’m just a journalist - we asked Dario’s own product if he’s lying through his teeth, and here’s what it said:”

geraneum9• 1 year ago

I’d love a journalist that do their job. For example when someone like this CEO pulls a number out of their ass, maybe push them on how they arrived at this? Why does it displace 50%? Why 70? Why not 45?

elktown• 1 year ago

Tech has a big problem of selective critical thinking due to a perpetual gold rush causing people to adopt a stockbroker mentality of not missing out on the next big thing - be it the next subfield like AI, the next cool tech that you can be an early adopter on etc. But yeah, nothing new under the sun; it's corruption.

mjburgess5• 1 year ago

In many spheres today "thought leadership" is a kind of marketing and sales activity. It is no wonder then that no one can think and no one can lead: either would be an fatal to healthy sales.

leeroihe3• 1 year ago

I used to be a big proponent of AI tools and llms, even built products around them. But to be honest, with all of the big AI ceos promising that they're going to "replace all white collar jobs" I can't see that they want what's best for the country or the american people. It's legitimately despicable and ghoulish that they just expect everyone to "adapt" to the downstream affects of their knowledge-machine lock-in.

bawana2• 1 year ago

When are going to get AI CEOs as a service?

bawana2• 1 year ago

When are we going to get AI CEOs as a service?

crims0n• 1 year ago

You may be onto something… sell strategic decisions by an AI cohort as a service, insure against the inevitable duds, profit.

0x5f3759df-i• 1 year ago

I asked ChatGPT to be a CEO and decide if everyone should work in office 5 days a week:

“ Final Thought (as a CEO):

I wouldn’t force a full return unless data showed a clear business case. Culture, performance, and employee sentiment would all guide the decision. I’d rather lead with transparency, flexibility, and trust than mandates that could backfire.

Would you like a sample policy memo I’d send to employees in this scenario?”

A better, more reasonable CEO than the one I have. So I’m looking forward to AI taking that white collar job especially.

CKMo• 1 year ago

There's definitely a big problem with entry-level jobs being replaced by AI. Why hire an intern or a recent college-grad when they lack both the expertise and experience to do what an AI could probably do?

Sure, the AI might require handholding and prompting too, but the AI is either cheaper or actually "smarter" than the young person. In many cases, it's both. I work with some people who I believe have the capacity and potential to one day be competent, but the time and resource investment to make that happen is too much. I often find myself choosing to just use an AI for work I would have delegated to them, because I need it fast and I need it now. If I handed it off to them I would not get it fast, and I would need to also go through it with them in several back-and-forth feedback-review loops to get it to a state that's usable.

Given they are human, this would push back delivery times by 2-3 business days. Or... I can prompt and handhold an AI to get it done in 3 hours.

Not that I'm saying AI is a god-send, but new grads and entry-level roles are kind of screwed.

mjburgess5• 1 year ago

This is always the case though. A factor of 50x productivity between expert and novice is small. Consider how long it take you to conduct foot surgery vs. a food surgeon -- close to a decade of medical school + medical experience -- just for a couple hours of work.

There have never been that many businesses able to hire novices for this reason.

pc86• 1 year ago

This is a big part of why a lot of developers' first 1-3 jobs are small mom & pop shops of varying levels of quality, almost none of which have "good" engineering cultures. Market rate for a new grad dev might be X, it's hard to find an entry level job at X but mom & pop business who needs 0.7 FTE developers is willing to pay 0.8X and even though the owner is batshit insane it's not a bad deal for the 22 and 23 year olds willing to do it.

InitialLastName• 1 year ago

I think for a surgeon as an example, quality may be a better metric than time. I'll bet I could conduct an attempted foot surgery way faster than a foot surgeon, but they're likely to conduct successful foot surgeries.

abletonlive8• 1 year ago

This is a big issue in the short term but in the long term I actually think AI is going to be a huge democratization of work and company building.

I spend a lot of time encouraging people to not fight the tide and spend that time intentionally experimenting and seeing what you can do. LLMs are already useful and it's interesting to me that anybody is arguing it's just good for toy applications. This is a poisonous mindset and results in a potentially far worse outcome than over-hyping AI for an individual.

I am wondering if I should actually quit a >500K a year job based around LLM applications and try to build something on my own with it right now.

I am NOT someone that thinks I can just craft some fancy prompt and let an LLM agent build me a company, but I think it's a very powerful tool when used with great intention.

The new grads and entry level people are scrappy. That's why startups before LLMs liked to hire them. (besides being cheap, they are just passionate and willing to make a sacrifice to prove their worth)

The ones with a lot of creativity have an opportunity right now that many of us did not when we were in their shoes.

In my opinion, it's important to be technically potent in this era, but it's now even more important to be creative - and that's just what so many people lack.

Sitting in front of a chat prompt and coming up with an idea is hard for the majority of people that would rather be told what to do or what direction to take.

My message to the entry-level folks that are in this weird time period. It's tough, and we can all acknowledge that - but don't let cynicism shackle you. Before LLMs, your greatest asset was fresh eyes and the lack of cynicism brought upon by years of industry. Don't throw away that advantage just because the job market is tough. You, just like everybody else, have a very powerful tool and opportunity right in front of you.

The amount of people trying to convince you that it's just a sham and hype means that you have less competition to worry about. You're actually lucky there's a huge cohort of experienced people that have completely dismissed LLMs because they were too egotistical to spend meaningful time evaluating it and experimenting with it. LLM capabilities are still changing every 6 months-1 year. Anybody that has decided concretely that there is nothing to see here is misleading you.

Even in the current state of LLM if the critics don't see the value and how powerful it is mostly a lack of imagination that's at play. I don't know how else to say it. If I'm already able to eliminate someone's role by using an LLM then it's already powerful enough in its current state. You can argue that those roles were not meaningful or important and I'd agree - but we as a society are spending trillions on those roles right now and would continue to do so if not for LLMs

izabera• 1 year ago

what does "huge democratization of work" even mean? what world do you people live in? the current global unemployment rate on my planet is around 5% so that seems pretty democratised already?

mechagodzilla• 1 year ago

Interns and new grads have always been a net-negative productivity-wise in my experience, it's just that eventually (after a small number of months/years) they turn into extremely productive more-senior employees. And interns and new grads can use AI too. This feels like asking "Why hire junior programmers now that we have compilers? We don't need people to write boring assembly anymore." If AI was genuinely a big productivity enhancer, we would just convert that into more software/features/optimizations/etc, just like people have been doing with productivity improvements in computers and software for the last 75 years.

alephnerd6• 1 year ago

It's a monetary issue at the end of the day.

AI/ML and Offshoring/GCCs are both side effects of the fact that American new grad salaries in tech are now in the $110-140k range.

At $70-80k the math for a new grad works out, but not at almost double that.

Also, going remote first during COVID for extended periods proved that operations can work in a remote first manner, so at that point the argument was made that you can hire top talent at American new grad salaries abroad, and plenty of employees on visas were given the option to take a pay cut and "remigrate" to help start a GCC in their home country or get fired and try to find a job in 60 days around early-mid 2020.

The skills aspect also played a role to a certain extent - by the late 2010s it was getting hard to find new grads who actually understood systems internals and OS/architecture concepts, so a lot of jobs adjacent to those ended up moving abroad to Israel, India, and Eastern Europe where universities still treat CS as engineering instead of an applied math disciple - I don't care if you can prove Dixon's factorization method using induction if you can't tell me how threading works or the rings in the Linux kernel.

The Japan example mentioned above only works because Japanese salaries in Japan have remained extremely low and Japanese is not an extremely mainstream language (making it harder for Japanese firms to offshore en masse - though they have done so in plenty of industries where they used to hold a lead like Battery Chemistry).

lokar3• 1 year ago

Where I have worked new grads (and interns) were explicitly negative.

This is part of why some companies have minimum terminal levels (often 5/Sr) before which a failure to improve means getting fired.

0xpgm3• 1 year ago

Isn't that every new employee? The first few months you are not expected to be firing on all cylinders as you catch up and adjust to company norms

An intern is much more valuable than AI in the sense that everyone makes micro decisions that contribute to the business. An Intern can remember what they heard in a meeting a month ago or some important water-cooler conversation and incorporate that in their work. AI cannot do that

brookst4• 1 year ago

I just can't agree with this argument at all.

Today, you hire an intern and they need a lot of hand-holding, are often a net tax on the org, and they deliver a modest benefit.

Tomorrow's interns will be accustomed to using AI, will need less hand-holding, will be able to leverage AI to deliver more. Their total impact will be much higher.

The whole "entry level is screwed" view only works if you assume that companies want all of the drawbacks of interns and entry level employees AND there is some finite amount of work to be done, so yeah, they can get those drawbacks more cheaply from AI instead.

But I just don't see it. I would much rather have one entry level employee producing the work of six because they know how to use AI. Everywhere I've worked, from 1-person startup to the biggest tech companies, has had a huge surplus of work to be done. We all talk about ruthless prioritization because of that limit.

So... why exactly is the entry level screwed?

gerad• 1 year ago

They don't have the experience to tell bad AI responses from good ones.

einpoklum3• 1 year ago

will need less hand-holding, will be able to leverage AI to deliver more

Well, maybe it'll be the other way around: Maybe they'll need more hand-holding since they're used to relying on AI instead of doing things themselves, and when faced with tasks they need to do, they will be less able.

But, eh, what am I even talking about? The _senior_ developers in a many companies need a lot of hand-holding that they aren't getting, write bad code, with poor practices, and teach the newbies how to get used to doing that. So that's why the entry-level people are screwed, AI or no.

chongli• 1 year ago

Tomorrow's interns will be accustomed to using AI, will need less hand-holding, will be able to leverage AI to deliver more.

Maybe tomorrow's interns will be "AI experts" who need less hand-holding, but the day after that will be kids who used AI throughout elementary school and high school and know nothing at all, deferring to AI on every question, and have zero ability to tell right from wrong among the AI responses.

I tutor a lot of high school students and this is my takeaway over the past few years: AI is absolutely laying waste to human capital. It's completely destroying students' ability to learn on their own. They are not getting an education anymore, they're outsourcing all their homework to the AI.

aloknnikhil• 1 year ago

It's not that entry-level jobs / interns are irrelevant. It's more that entry-level has been redefined and it requires significant uplevelling in terms of skills necessary to do a job at that level. That's not necessarily a bad thing. As others have said here, I would be more willing to hand-off more complex tasks to interns / junior engineers because my expectation is they leverage AI to tackle it faster and learn in the process.

ChrisMarshallNY4• 1 year ago

This is where the horrific disloyalty of both companies and employees, comes to bite us in the ass.

The whole idea of interns, is as training positions. They are supposed to be a net negative.

The idea is that they will either remain at the company, after their internship, or move to another company, taking the priorities of their trainers, with them.

But nowadays, with corporate HR, actively doing everything they can to screw over their employees, and employees, being so transient, that they can barely remember the name of their employer, the whole thing is kind of a worthless exercise.

At my old company, we trained Japanese interns. They would often relocate to the US, for 2-year visas, and became very good engineers, upon returning to Japan. It was well worth it.

FirmwareBurner5• 1 year ago

At my old company, we trained Japanese interns. They would often relocate to the US, for 2-year visas, and became very good engineers,

Damn, I wish that was me. Having someone mentor you at the beginning of your career instead of having to self learn and fumble your way around never knowing if you're on the right track or not, is massive force multiplier that pays massive dividends over your career. It's like entering the stock market with 1 million $ capital vs 100 $. You're also less likely to build bad habits if nobody with experience teaches you early on.

geraneum9• 1 year ago

horrific disloyalty of both companies and employees

There’s no such a thing as loyalty in employer-employee relationships. There’s money, there’s work and there’s [collective] leverage. We need to learn a thing or two from blue collars.

neilv• 1 year ago

I agree that interns are pretty much over in tech. Except maybe for an established company do do as a semester/summer trial/goodwill period, for students near graduation. You usually won't get work output worth the mentoring cost, but you might identify a great potential hire, and be on their shortlist.

Startups are less enlightened than that about "interns".

Literally today, in a startup job posting, to a top CS department, they're looking for "interns" to bring (not learn) hot experience developing AI agents, to this startup, for... $20/hour, and get called an intern.

It's also normal for these startup job posts to be looking for experienced professional-grade skills in things like React, Python, PG, Redis, etc., and still calling the person an intern, with a locally unlivable part-time wage.

Those startups should stop pretending they're teaching "interns" valuable job skills, admit that they desperately need cheap labor for their "ideas person" startup leadership, to do things they can't do, and cut the "intern" in as a founding engineer with meaningful equity. Or, if you can't afford to pay a livable and plausibly competitive startup wage, maybe they're technical cofounders.

xpe2• 1 year ago

I personally care a lot about people, but if I was running a publicly traded for-profit, I would have a lot of constraints about how to care for them. (A good place to start, by the way, is not bullshitting people about the financial realities.)

Employees are lucky when incentives align and employers treat them well. This cannot be expected or assumed.

A lot of people want a different kind of world. If we want it, we’re gonna have to build it. Think about what you can do. Have you considered running for office?

I don’t think it is helpful for people to play into the victim narrative. It is better to support each other and organize.

Loughla• 1 year ago

So what happens when you retire and have no replacement because you didn't invest in entry level humans?

This feels like the ultimate pulling up the ladder after you type of move.

dylan6043• 1 year ago

Are you honestly trying to tell us that the code you receive from an AI is not requiring any of your time to review and tweak and is 100% correct every time and ready to deploy into your code base with no changes what so ever? You my friend must be a steely eyed missile man of prompting

throwuxiytayq• 1 year ago

Consider that there are no humans in existence that fulfill your requirements, not to mention $20/mo ones

diogolsq• 1 year ago

You’re right that AI is fast and often more efficient than entry-level humans for certain tasks — but I’d argue that what you’re describing isn’t delegation, it’s just choosing to do the work yourself via a tool. Implementation costs are lower now, so you decide to do it on your own.

Delegation, properly defined, involves transferring not just the task but the judgment and ownership of its outcome. The perfect delegation is when you delegate to someone because you trust them to make decisions the way you would — or at least in a way you respect and understand.

You can’t fully delegate to AI — and frankly, you shouldn’t. AI requires prompting, interpretation, and post-processing. That’s still you doing the thinking. The implementation cost is low, sure, but the decision-making cost still sits with you. That’s not delegation; it’s assisted execution.

Humans, on the other hand, can be delegated to — truly. Because over time, they internalize your goals, adapt to your context, and become accountable in a way AI never can.

Many reasons why AI can't fill your shoes:

1. Shallow context – It lacks awareness of organizational norms, unspoken expectations, or domain-specific nuance that’s not in the prompt or is not explicit in the code base.

2. No skin in the game – AI doesn’t have a career, reputation, or consequences. A junior human, once trained and trusted, becomes not only faster but also independently responsible.

Junior and Interns can also use AI tools.

dasil003• 1 year ago

You said exactly what I came here to say.

Maybe some day AI will truly be able to think and reason in a way that can approximate a human, but we're still very far from that. And even when we do, the accountability problem means trusting AI is a huge risk.

It's true that there are white collar jobs that don't require actual thinking, and those are vulnerable, but that's just the latest progression of computerization/automation that's been happening steadily for the last 70 years already.

It's also true that AI will completely change the nature of software development, meaning that you won't be able to coast just on arcane syntax knowledge the way a lot of programmers have been able to so far. But the fundamental precision of logical thought and mapping it to a desirable human outcome will still be needed, the only change is how you arrive there. This actually benefits young people who are already becoming "AI native" and will be better equipped to leverage AI capabilities to the max.

einpoklum3• 1 year ago

Why hire an intern or a recent college-grad when they lack both the expertise and experience to do what an AI could probably do?

1. Because, generally, they don't.

2. Because an LLM is not a person, it's a chatbot.

3. "Hire an intern" is that US thing when people work without getting real wages, right?

Grrr :-(

aianus2• 1 year ago

Interns make $75k+ in tech in the US. It's definitely not unpaid. In fact my school would not give course credit for internships if they were unpaid.

baxtr• 1 year ago

I think it’s the other way around.

If LLMs continue to become more powerful, hiring more juniors who can use them will be a no-brainer.

phatfish• 1 year ago

Yup, apart from a few companies at the cutting edge the most difficult problems to solve in a work environment are not technical.

phailhaus• 1 year ago

I don't get this because someone has to work with the AI to get the job done. Those are the entry-level roles! The manager who's swamped with work sure as hell isn't going to do it.

necheffa• 1 year ago

Why hire an intern or a recent college-grad when they lack both the expertise and experience to do what an AI could probably do?

AI can barely provide the code for a simple linked list without dropping NULL pointer dereferences every other line...

Been interviewing new grads all week. I'd take a high performing new grad that can be mentored into the next generation of engineer any day.

If you don't want to do constant hand holding with a "meh" candidate...why would you want to do constant hand holding with AI?

I often find myself choosing to just use an AI for work I would have delegated to them, because I need it fast and I need it now.

Not sure what you are working on. I would never prioritize speed over quality - but I do work in a public safety context. I'm actually not even sure of the legality of using an AI for design work but we have a company policy that all design analysis must still be signed off on by a human engineer in full as if it were 100% their own.

I certainly won't be signing my name on a document full of AI slop. Now an analysis done by a real human engineer with the aid of AI - sure, I'd walk through the same verification process I'd walk through for a traditional analysis document before signing my name on the cover sheet. And that is something a jr. can bring to me to verify.

pedalpete• 1 year ago

We've been doing the exact opposite for some positions.

I've been interviewing marketing people for the last few months (I have a marketing background from long ago), and the senior people were either way too expensive for our bootstrapped start-up, or not of the caliber we want in the company.

At the same time, there are some amazing recent grads and even interns who can't get jobs.

We've been hiring the younger group, and contracting for a few days a week with the more experienced people.

Combine that with AI, and you've got a powerful combination. That's our theory anyway.

It's worked pretty well with our engineers. We are a team of 4 experienced engineers, though as CEO I don't really get to code anymore, and 1 exceptional intern. We've just hired our 2nd intern.

jmyeet• 1 year ago

This is basically what happened after 2008. The entry level jobs college grads did basically disappeared and didn't really come back for many years. So we kind of lost half a generation. Those who missed out are the ones who weren't able to buy a house or start a family and are now in their 40s, destined to be permanent renters who can never retire.

The same thing will happen to Gen Z because of AI.

In both cases, the net effect of this (and the desired outcome) is to suppress wages. Not only of entry-level job but every job. The tech sector is going to spend the next decade clawing back the high costs of tech people from the last 15-20 years.

The hubris here is that we've had a unprecedented boom such that many in the workforce have never experienced a recession, what I'd call "children of summer" (to borrow a George RR Martin'ism). People have fallen into the trap of the myth of meritocracy. Too many people thing that those who are living paycheck to paycheck (or are outright unhoused) are somehow at fault when spiralling housing costs, limited opportunities and stagnant real wages are pretty much responsible for everything.

All of this is a giant wealth transfer to the richest 0.01% who are already insanely wealthy. I'm convinced we're beyond the point where we can solve the problems of runaway capitalism with electoral politics. This only ends in tyranny of a permanent underclass or revolution.

hollowturtle• 1 year ago

imo comparing entry-level people with ai is very short sighted, I was smarter than every dumb dinosaur at my first job, I was so eager to learn and proactive and positive... i probably was very lucky too but my point is i don't believe this whole thing that a junior is worse than ai, i'd rather say the contrary

sauercrowd• 1 year ago

"intern" and "entry level" are proxies for complexity with these comparisons, not actual seniority. We'll keep hiring interns and entry level positions, they'll just do other things.

snowwrestler3• 1 year ago

Companies reducing young hires because of AI are doing it backward. Returns on AI will be accelerated by early-career staff because they are already eagerly using AI in daily life, and have the least attachment to how jobs are done now.

You’re probably not going to transform your company by issuing Claude licenses to comfortable middle-aged career professionals who are emotionally attached to their personal definition of competency.

Companies should be grabbing the kids who just used AI to cheat their way through senior year, because that sort of opportunistic short-cutting is exactly what companies want to do with AI in their business.

sarchertech5• 1 year ago

If the AI can write code to a level that doesn’t need an experienced person to check the output, you don’t need tech companies at all.

anshumankmr• 1 year ago

Not that I'm saying AI is a god-send, but new grads and entry-level roles are kind of screwed.

A company that I know of is having a L3 hiring freeze also and some people are downgraded from L4 to L3 or L5 to L4 also.. Getting more work for less cost.

uludag• 1 year ago

I thought the whole idea of automation though was to lower the skill requirement. Everyone compares AI to the industrial revolution and the shift from artisan work to factory work. If this analogy were to hold true, then what employers should actually be wanting is more junior devs, maybe even non-devs, hired at a much cheaper wage. A senior dev may be able to outperform a junior by a lot, but assuming the AI is good enough, four juniors or like 10 non-devs should be able to outperform a senior.

This obviously not being the case shows that we're not in a AI driven fundamental paradigm shift, but rather run of the mill cost cutting measures. Like suppose a tech bubble pops and there are mass layoffs (like the Dotcom bubble). Obviously people will loose their jobs. AI hype merchants will almost definitely try to push the narrative that these losses are from AI advancements in an effort to retain funding.

runeks2• 1 year ago

Sure, the AI might require handholding and prompting too, but the AI is either cheaper or actually "smarter" than the young person.

The AI will definitely require handholding. And that hand-holder will be an intern or a recent college-grad.

Animats• 1 year ago

The real bloodbath will come when coordination between multiple AIs, in a company sense, starts working. Computers have much better I/O than humans. Once a corporate organization can be automated, it will be too fast for humans to participate. There will be no place for slow people.

"Move fast and break things" - Zuckerberg

"A good plan violently executed now is better than a perfect plan executed next week." - George S. Patton

catigula11• 1 year ago

This doesn't even make sense. What corporations do you think will exist in this world?

You're not going to sell me your SaaS when I can rent AIs to make faster cheaper IP that I actually own to my exact specifications.

ofjcihen4• 1 year ago

This is always the indicator I look for whether or not someone actually knows what they’re talking about.

If you can’t extrapolate on your own thesis you can’t be knowledgeable in the field.

Good example was a guy on here who was convinced every company would be ran by one person because of AI. You’d wake up in the morning and decide which products your AI came up with while you slept would be profitable. The obvious next question is “then why are you even involved?”

sbierwagen• 1 year ago

https://slatestarcodex.com/2016/05/30/ascended-economy/

Robot run iron mine that sells iron ore to a robot run steel mill that sells steel plate to a robot run heavy truck manufacturer that sells heavy trucks to robot run iron mines, etc etc.

The material handling of heavy industry is already heavily automated, almost by definition. You just need to take out the last few people.

ck2• 1 year ago

LLM is going to be used for oppression by every government, not just dictatorships but USA of course

Think of it as an IQ test of how new technology is used

Let me give you an easier example of such a test

Let's say they suddenly develop nearly-free unlimited power, ie. fusion next year

Do you think the world will become more peaceful or much more war?

If you think peaceful, you fail, of course more war, it's all about oppression

It's always about the few controlling the many

The "freedom" you think you feel on a daily basis is an illusion quickly faded

deadbabe2• 1 year ago

Something I’ve come to realize in the software industry is: if you have more smart engineers than the competition, you win.

If you don’t snatch up the smartest engineers before your competition does: you lose.

Therefore at a certain level of company, hiring is entirely dictated by what the competition is doing. If everyone is suddenly hiring, you better start doing it too. If no one is, you can relax, but you could also pull ahead if you decide to hire rapidly, but this will tip off competitors and they too will begin hiring.

Whether or not you have any use for those engineers is irrelevant. So AI will have little impact on hiring trends in this market. The downturn we’ve seen in the past few years is mostly driven by the interest rate environment, not because AI is suddenly replacing engineers. An engineer using AI gives more advantage than removing an engineer, and hiring an engineer who will use AI is more advantageous than not hiring one at all.

AI is just the new excuse for firing or not hiring people, previously it was RTO but that hype cycle has been squeezed for all it can be.

ggm• 1 year ago

Without well paid middle classes, who is buying all the fancy goods and services?

Money is just rationing. If you devalue the economy implicitly you accept that, and the consequences for society at large.

Lenin's dictum: A capitalist will sell you the rope you hang him with Comes to mind

Hilift2• 1 year ago

Without well paid middle classes, who is buying all the fancy goods and services?

People charging on their credit cards. Consumers are adding $2 billion in new debt every day.

"Total household debt increased by $167 billion to reach $18.20 trillion in the first quarter"

https://www.newyorkfed.org/microeconomics/hhdc

ramesh312• 1 year ago

Without well paid middle classes, who is buying all the fancy goods and services?

Rich people buying even fancier goods and services. You already see this in the auto industry. Why build a great $20,000 car for the masses when you can make the same revenue selling $80,000 cars to rich people (and at higher margins)? This doesn't work of course when you have a reasonably egalitarian society with reasonable wealth inequality. But the capitalists have figured out how to make 75% of us into willing slaves for the rest. A bonus of this is that a good portion of that 75% can be convinced to go into lifelong debt to "afford" those things they wish they could actually buy, further entrenching the servitude.

WaltPurvis• 1 year ago

I plugged those two quotes from Amodei into ChatGPT along with this prompt: "Pretend you are highly skeptical about the potential of AI, both in general and in its potential for replacing human workers the way Amodei predicts. Write a quick 800-word takedown of his predictions."

I won't paste in the result here, since everyone here is capable of running this experiment themselves, but trust me when I say ChatGPT produced (in mere seconds, of course) an article every bit as substantive and well-written as the cited article. FWIW.

trhway2• 1 year ago

Read on about PLTR in recent days - all these government layoffs (including by DOGE well connected to PLTR) with the money redirected toward the Grand Unification Project using PLTR Foundry (with AI) platform.

chris_armstrong• 1 year ago

The wildest claims are those of increased labor productivity and economic growth: if they were true, our energy consumption would be increasing wildly beyond our current capacity to add more (dwarfing the increase from AI itself).

Productivity doesn’t increase on its own; economists struggle to separate it from improved processes or more efficient machinery (the “multi factor productivity fudge”). Increased efficiency in production means both more efficient energy use AND being able to use a lot more of it for the same input of labour.

smeeger• 1 year ago

if being redundant would lead to mass layoffs then half of white collar workers would have been laid off decades ago. and white collar people will fiddle with rules and regulations to make their ever more bloated redundancy even more brazen with the addition of AI… and then later when AI has the ability to replace blue collar workers it will do so immediately and swiftly while the white collar people get all the money. its happened a thousand times before and will happen again.

stephc_int13• 1 year ago

The main culprit behind the hype of the AI revolution is a lack of understanding of its true nature and capabilities. We should know better, Eliza demonstrated decades ago how easily we can be fooled by language, this is different and more useful but we rely so much on language fluency and knowledge retrieval as a proxy for intelligence that we are fooled again.

I am not saying this is a nothing burger, the tech can be applied to many domains and improve productivity, but it does not think, not even a little, and scaling won’t make that magically happen.

Anyone paying attention should understand this fact by now.

There is no intelligence explosion in sight, what we’ll see during the next few years is a gradual and limited increase in automation, not a paradigm change, but the continuation of a process that started with the industrial revolution.

gcanyon2• 1 year ago

...everyone here saying "someday AI will <fill in the blank> but not today" while failing to acknowledge that for a lot of things "someday" is 2026, and for an even larger number of things it's 2027, and we can't even predict whether or not in 2028 AI will handle nearly all things...

causal2• 1 year ago

The problem is that it's hard to pin down any job that's been eliminated by AI even after years of having LLMs. I'm sure it will happen. It just seems like the trajectory of intelligence defies any simple formula.

gcanyon2• 1 year ago

There's definitely an element of what we saw in the '90s -- software didn't always make people faster, it made the quality of their output better (wysiwyg page layout, better database tools/validation, spell check in email, etc. etc.).

But we're going to get to a point where "the quality goes up" means the quality exceeds what I can do in a reasonable time frame, and then what I can do in any time frame...

sfblah2• 1 year ago

I literally am in the process of firing someone who we no longer need because of efficiencies tied to GenAI. I work at a top-10 tech company. So, there you go. That's one job.

CSMastermind• 1 year ago

Huge amounts of white collar jobs have been automated since the advent of computers. If you look at the work performed by office workers in the 1960s and compared it to what people today do it'd be almost unrecognizable.

They spent huge amounts of time on things that software either does automatically or makes 1,000x faster. But by and large that actually created more white collar jobs because those capabilities meant more was getting done which meant new tasks needed to be performed.

janalsncm• 1 year ago

I don’t like this argument because 1) it doesn’t address the social consequences of rapid onset and large scale unemployment and 2) there is no law of nature that a job lost here creates a new job there.

On the first point, unemployment during the Great Depression was “only” 30%. And those people were eventually able to find other jobs. Here, we are talking about permanent unemployment for even larger numbers of people.

The Luddites were right. Machines did take their jobs. Those individuals who invested significantly in their craft were permanently disadvantaged. And those who fought against it were executed.

And on point 2, to be precise, a lack of jobs doesn’t mean a lack of problems. There are a ton of things society needs to have accomplished, and in a perfect world the guy who was automated out of packing Amazon boxes could open a daycare for low income parents. We just don’t have economic models to enable most of those things, and that’s only going to get worse.

ryukoposting• 1 year ago

I'll preface this by saying I agree with most of what you said.

It'll be a slow burn, though. The projection of rapid, sustained large-scale unemployment assumes that the technology rapidly ascends to replace a large portion of the population at once. AI is not currently on a path to replacing a generalized workforce. Call center agents, maybe.

Second, simply "being better at $THING" doesn't mean a technology will be adopted, let alone quickly. If that were the case, we'd all have Dvorak keyboards and commuter rail would be ubiquitous.

Third, the mass unemployment situation requires economic conditions where not leveraging a presumably exploitable underclass of unemployed persons is somehow the most profitable choice for the captains of industry. They are exploitable because this is not a welfare state, and our economic safety net is tissue-paper thin. We can, therefore, assume their labor can be had at far less than its real worth, and thus someone will find a way to turn a profit off it. Possibly the Silicon Valley douchebags who caused the problem in the first place.

ccorcos13• 1 year ago

What makes you so concerned about rapid onset of we haven’t seen any significant change in the (USA) unemployment rate?

And there are some laws of nature that are relevant such as supply-demand economics. Technology often makes things cheaper which unlocks more demand. For example, I’m sure many small businesses would love to build custom software to help them operate but it’s too expensive.

anthomtb• 1 year ago

Huge amounts of white collar jobs have been automated since the advent of computers

One of which was the occupation of being a computer!

lambdasquirrel• 1 year ago

Anecdotal, but AI was what enabled me to learn French, when I was doing that. Before LLMs, I would've had to pay a lot more money to get the class time I'd need, but the availability of Google Translate and DeepL meant that some meaningful, casual learning was within reach. I could reasonably study, try to figure things out, and have questions for the teachers the two or three times a week I had lessons.

Nowadays I'm learning my parents' tongue (Cantonese) and Mandarin. It's just comical how badly the LLMs do sometimes. I swear they roll a natural 1 on a d20 and then just randomly drop a phrase. Or at least that's my head canon. They're just playing DnD on the side.

PeterHolzwarth2• 1 year ago

The classic example is the 50's/60's photograph of an entire floor of a tall office building replaced by single spreadsheet. This passed without comment.

joshdavham• 1 year ago

This type of hype is pretty perplexing to me.

Supposing that you are trying to increase AI adoption among white-collar workers, why try to scare the shit out them in the process? Or is he moreso trying to sell to the C-suite?

taormina• 1 year ago

He’s selling exclusively to the C-suite. Why would he care about the white collar workers? He wouldn’t be trying to put them all out of work if he cares

chr15m• 1 year ago

Because it creates FOMO which creates sales.

globalnode• 1 year ago

i really liked this article, it puts into perspective how great claims require great proof, and so far all we've heard are great claims. i love ml tech but i just dont trust it to replace a human completely. sure it can augment roles but thats not the vision we're being sold.

osigurdson2• 1 year ago

The real value is going to be in areas that neither machines nor humans could do previously.

snowwrestler3• 1 year ago

Historically, people have been pretty good at predicting the effects of new technologies on existing jobs. But quite bad at predicting the new jobs / careers / industries that are eventually created with those technologies.

This is why free market economies create more wealth over time than centrally planned economies: the free market allows more people to try seemingly crazy ideas, and is faster to recognize good ideas and reallocate resources toward them.

In the absence of reliable prediction, quick reaction is what wins.

Anyway, even if AI does end up “destroying” tons of existing white collar jobs, that does not necessarily imply mass unemployment. But it’s such a common inference that it has its own pejorative: Luddite.

And the flip side of Ludddism is what we see from AI boosters now: invoking a massive impact on current jobs as a shorthand to create the impression of massive capability. It’s a form of marketing, as the CNN piece says.

beepbooptheory• 1 year ago

So, we are doomed to work forever, just maybe different jobs?

absurdo3• 1 year ago

Basically yeah. You live in a world of layered servitude and, short of a financial windfall that hoists you up for some time, you’re basically guaranteed to work your entire life, and grow old, frail and poor. This isn’t a joke, it’s reality for many people that’s hidden from us to keep us deluded. Similar to my other mini-rant, I don’t have any valid answers to the problem at hand. Just acknowledging how fucked things are for humanity.

satvikpendem3• 1 year ago

Of course. I mean this has never not been the case unless you are independently wealthy. Work always expands, that's why it's a fallacy to think that if we just had more productivity gains that we'd work half the time; no, there are always new things to do tomorrow that were not possible yesterday.

77pt77• 1 year ago

Just like the Red Queen.

You have to always keep on moving just to stay in the same place.

tw04• 1 year ago

But also it potentially means mass unemployment and we have literally no plan in place if that happens beyond complete societal collapse.

Even if you think all the naysayers are “luddites”, do you really think it’s a great idea to have no backup plan beyond “whupps we all die or just go back to the Stone Age”?

ccorcos13• 1 year ago

do you really think it’s a great idea to have no backup plan

What makes you think people haven’t made back up plans?

Or are you saying government needs to do it for us?

snowwrestler3• 1 year ago

We actually have many backup plans. The most effective ones will be the new business plans that unlock investment which is what creates new jobs. But behind that are a large set of government policies and services that help people who have lost work. And behind that are private resources like charities, nonprofits, even friends and family.

People don’t want society to collapse. So if you think it’s something that people can prevent, feel comforted that everyone is trying to prevent it.

nopinsight• 1 year ago

My thesis is that this could lead to a booming market for “pink-collar” service jobs. A significant latent demand exists for more and better services in developed countries.

For instance, upper-middle-class and middle-class individuals in countries like India and Thailand often have access to better services in restaurants, hotels, and households compared to their counterparts in rich nations.

Elderly care and health services are two particularly important sectors where society could benefit from allocating a larger workforce.

Many others will have roles to play building, maintaining, and supervising robots. Despite rapid advances, they will not be as dexterous, reliable, and generally capable as adult humans for many years to come. (See: Moravec's paradox).

digdugdirk• 1 year ago

More people need to understand the actual history of the luddites. The real issue was the usage of mechanized equipment to overwhelm an entire sector of the economy of the day - destroying the labor value of a vast swath of craftspeople and knocking them down a peg on the social ladder.

Those people who were able to get work were now subject to a much more dangerous workplace and forced into a more rigid legalized employer/employee structure, which was a relatively new "corporate innovation" in the grand scheme of things. This, of course, allowed/required the state to be on the hook for enforcement of the workplace contract, and you can bet that both public and private police forces were used to enforce that contract with violence.

Certainly something to think about for all the users on this message board who are undoubtedly more highly skilled craftspeople than most, and would never be caught up in a mass economic displacement driven by the introduction of a new technological innovation.

At the very least, it's worth a skim through the Wikipedia article: https://en.wikipedia.org/wiki/Luddite

csomar3• 1 year ago

I think the takeaway is that interest rates have to be maintained relatively high as the ZIRP era has showed that it breaks the free market. There is a reason why the Trump wants to lower the interest rate.

Sure it is painful but a ZIRP economy doesn't listen to the end consumers. No reason to innovate and create crazy ideas if you have plenty of income.

madaxe_again6• 1 year ago

When steam engines came along, an awful lot of people argued that being able to pump water from mines faster, while inarguably useful, would not have any broad economical impact. Only madmen saw the Newcomen engine and thought “ah, railways!”. Those madmen became extraordinarily wealthy. Vast categories of work were eliminated, others were created.

I think this situation is very similar in terms of the underestimation of scope of application, however differs in the availability of new job categories - but then that may be me underestimating new categories which are as yet as unforeseen as stokers and train conductors once were.

qgin3• 1 year ago

I often see people say “AI can’t do ALL of my job, so that means my job is safe.

But what this means at scale, over time, is that if AI can do 80% of your job, AI will do 80% of your job. The remaining 20% human-work part will be consolidated and become the full time job of 20% of the original headcount while the remaining 80% of the people get fired.

AI does not need to do 100% of any job (as that job is defined today ) to still result in large scale labor reconfigurations. Jobs will be redefined and generally shrunk down to what still legitimately needs human work to get it done.

As an employee, any efficiency gains you get from AI belong to the company, not you.

sram1337• 1 year ago

...or your job goes from commanding a $200k/yr salary to $60k/yr. Hopefully that's enough to pay your mortgage.

noisy_boy3• 1 year ago

I am more worried about the resultant breakdown of the social order due to the social pressures and hopelessness and it's ramifications. Which next-bonus-cycle CEOs don't give a shit about. UBI is just talk (which few are talking) and even that has issues.

tdeck5• 1 year ago

Maybe someone can help me wrap my head around this in a different way, because here's how I see it.

If these tools are really making people so productive, shouldn't it be painfully obvious in companies' output? For example, if these AI coding tools were an amazing productivity boost in the end, we'd expect to see software companies shipping features and fixes faster than ever before. There would be a huge burst in innovative products and improvements to existing products. And we'd expect that to be in a way that would be obvious to customers and users, not just in the form of some blog post or earnings call.

For cost center work, this would lead to layoffs right away, sure. But companies that make and sell software should be capitalizing on this, and only laying people off when they get to the point of "we just don't know what to do with all this extra productivity, we're all out of ideas!". I haven't seen one single company in this situation. So that makes me think that these decisions are hype-driven short term thinking.

CMCDragonkai2• 1 year ago

It's cause there are still bottlenecks. AI is definitely boosting productivity in specific areas, but the total system output is bottlenecked. I think we will see these bottlenecks get rerouted or refactored in the coming years.

tdeck5• 1 year ago

What do you think the main bottlenecks are right now?

esperent2• 1 year ago

It's cause there are still bottlenecks

How do you know this? What are the bottlenecks?

_heimdall• 1 year ago

AI is definitely boosting productivity in specific areas

What makes you so sure of the productivity boost when we aren't seeing a change in output?

jayd164• 1 year ago

We'll take cheaper over faster but is that the case? If it's not cheaper or faster what is the point?

AznHisoka2• 1 year ago

“we'd expect to see software companies shipping features and fixes faster than ever before. There would be a huge burst in innovative products and improvements to existing products.”

Shipping features faster != innovation or improvements to existing products

tdeck5• 1 year ago

Granting that those don't fully overlap, is that relevant to the point? I'm not seeing either.

epgui• 1 year ago

And?

ccorcos13• 1 year ago

AI tools seem to be most useful for little things. Fixing a little bug, making a little change. But those things aren’t always very visible or really move the needle.

It may help you build a real product feature quicker, but AI is not necessarily doing the research and product design which is probably the bottleneck for seeing real impact.

droopyEyelids• 1 year ago

If they're fixing all the little bugs that should give everyone much more time to think about product design and do the research.

wcfrobert2• 1 year ago

If AI makes everyone 10x engineers, you can 2x the productive output while reducing headcount by 5x.

Luckily software companies are not ball bearings factories.

tikhonj• 1 year ago

unluckily, too many corporate managers seem to think they are :/

danenania3• 1 year ago

If AI makes everyone 10x engineers, you can 2x the productive output while reducing headcount by 5x.

Why wouldn't you just 10x the productive output instead?

hoosieree3• 1 year ago

Or a 4-hour workweek.

bjt123453• 1 year ago

The problem seems to be two-fold.

Firstly, the capex is currently too high for all but the few.

This is a rather obvious statement, sure. But the impact is a lot of companies "have tried language models and they didn't work", and the capex is laughable.

Secondly, there's a corporate paralysis over AI.

I received a panicky policy statement written in legalaise forbidding employees from using LLMs in any form. Written both out of a panic regarding intellectual property leaking but also a panic about how to manage and control staff moving forward.

I think a lot of corporates still clutch at this view that AI will push the workforce costs down and are secretly wasting a lot money failing at this.

The waste is extraordinary, but it's other peoples money (it's actually the shareholders money) and it's seen as being all for a good cause and not something to discuss after it's gone. I can never get it discussed.

Meanwhile, at a grass roots level, I see AI is being embraced and is improving productivity, every second IT worker is using it, it's just that because of this corporate panicking and mismanagement, it's value is not yet measured.

tdeck5• 1 year ago

This is a good reminder that every org is different. However some companies like Microsoft are aggressively pushing AI tools internally, to a degree that is almost cringe.

bawolff8• 1 year ago

Firstly, the capex is currently too high for all but the few.
This is a rather obvious statement,

Nobody is saying companies have to make LLMs themselves.

SASS is a thing.

grumpymuppet• 1 year ago

The problem with this sort of analysis is that it's incremental and balanced across a large institution usually.

I think the reality is less like a switch and more like there are just certain jobs that get easier and you just need fewer people overall.

And you DO see companies laying off people in large numbers fairly regularly.

simonsarris3• 1 year ago

And you DO see companies laying off people in large numbers fairly regularly.

Sure but, so far, too regularly to be AI-gains-driven (at least in software). We have some data on software job postings and the job apocalypse, and corresponding layoffs, coincided with the end of ultra-low interest rates. If AI had a recent effect this year or last, its quite tiny in comparison.

https://fred.stlouisfed.org/graph/?g=1JmOr

so one can argue more is to come, but its hard to see how its had a real effect on jobs/layoffs thus far.

hyperadvanced2• 1 year ago

Layoffs happen because cash is scarce. In fact, cash is so scarce for anything that’s not “AI” that it’s basically nonexistent for startup fundraising purposes.

ivape7• 1 year ago

Companies are not accepting that their entire business will mostly go away. They are mostly frogs boiling in water, that's why they are kinda just incorporating these little chat bots and LLMs into their business, but the truth of the matter is it's all going away and it's impossible to believe. Take something like JIRA, it's entirely laughable because a simple LLM can handle entire project management with freaking voice with zero programming. They just don't believe that's the reality, we're talking about Kodak moment.

Worker productivity is secondary to business destruction, which is the primary event we're really waiting for.

nradov4• 1 year ago

That's silly. You still need a way to track and prioritize tasks even if you use voice input. Jira may be replaced with something better, built around an LLM from the ground up. But the basic project management requirements will never go away.

badsectoracula• 1 year ago

Take something like JIRA, it's entirely laughable because a simple LLM can handle entire project management with freaking voice with zero programming

When I used a not-so-simple LLM to make it act as a text adventure game it could barely keep track of the items in my inventory, so TBH i am a little bit skeptical that an LLM can handle entire project management - even without voice.

Perhaps it might be able to use tools/MCP/RPC to call out to real project management software and pretend to be your accountant/manager/whoever, but i wouldn't call that the LLM itself doing the project management task - and someone would need to write that project management software.

kraig911• 1 year ago

Effort in this equation isn't measured in man hours saved but dollars saved. We all know this is BS and isn't going to manifest this way. It's tantamount for giving framers a nailgun versus a hammer. We'll still be climbing the same rafters and doing the same work.

topspin2• 1 year ago

"shouldn't it be painfully obvious in companies' output?"

No.

The bottleneck isn't intellectual productivity. The bottleneck is a legion of other things; regulation, IP law, marketing, etc. The executive email writers and meeting attenders have a swarm of business considerations ricocheting around in their heads in eternal battle with each other. It takes a lot of supposedly brilliant thinking to safely monetize all the things, and many of the factors involved are not manifest in written form anywhere, often for legal reasons.

One place where AI is being disruptive is research: where researchers are applying models in novel ways and making legitimate advances in math, medicine and other fields. Another is art "creatives": graphic artists in particular. They're early victims and likely to be fully supplanted in the near future. A little further on and it'll be writers, actors, etc.

ImaCake• 1 year ago

Maybe this means that LLMs are ultimately good for small buisness. If large buisness is constrained by being large and LLMs are equally accesible to 5 people or 100 then surely what we will see is increased productivity in small companies?

bawolff8• 1 year ago

Even still, in theory this should free up more money to hire more lawyers, markerters, etc. The effect should still be there presuming the market isn't saturated with new ideas .

SteveNuts• 1 year ago

A little further on and it'll be writers, actors, etc.

The tech is going to have to be absolutely flawless, otherwise the uncanny-valley nature of AI "actors" in a movie will be as annoying as when the audio and video aren't perfectly synced in a stream. At least that's how I see it..

csomar3• 1 year ago

where researchers are applying models in novel ways and making legitimate advances in math, medicine and other fields.

Can you give an example, say in Medicine, where AI made a significant advancement? That is we are talking neural networks and up (ie: LLM) and not some local optimization.

throwaway20379• 1 year ago

    > One place where AI is being disruptive is research: where researchers are applying models in novel ways and making legitimate advances in math, medicine and other fields.

Great point. The perfect example: (From Wiki):

    > In 2024, Hassabis and John M. Jumper were jointly awarded the Nobel Prize in Chemistry for their AI research contributions for protein structure prediction.

AFAIK: They are talking about DeepMind AlphaFold.

Related: (Also from Wiki):

    > Isomorphic Labs Limited is a London-based company which uses artificial intelligence for drug discovery. Isomorphic Labs was founded by Demis Hassabis, who is the CEO.

pera• 1 year ago

Bullshit: Chatbots are not failing to demonstrate a tangible increase in companies' output because of regulations and IP law, they are failing because they are still not good for the job.

LLMs only exist because the companies developing them are so ridiculously powerful that can completely ignore the rule of law, or if necessary even change it (as they are currently trying to do here in Europe).

Remember we are talking about a technology created by torrenting 82 TB of pirated books, and that's just one single example.

"Steal all the users, steal all the music" and then lawyer up, as Eric Schmidt said at Stanford a few months ago.

throwawayffffas• 1 year ago

The things you mention in the legion of other things are actually things LLMs do better than intellectual productivity. They can spew entire libraries of marketing bs, summarize decades of legal precedents and fill out mountains of red tape checklists.

They have trouble with debugging obvious bugs though.

Teever• 1 year ago

Maybe in some industries and for some companies and their products but not all.

Like let's take operating systems as an example. If there are great productivity gains from LLMs while aren't companies like Apple, Google and MS shipping operating systems with vastly less bugs and cleaning up backlogged user feature requests?

gibbitz2• 1 year ago

Don't forget that to start a company the more common goal is to make money, not to produce a good or service that others want or need including employment. Meaning this is just a lesser concern. Generally business owners are just as likely to shed quality as to shed employees as long as the profits go up. What helps to sustain this is the gradual lowering of the bar on quality that leads to consumers settling for garbage products and sending a positive signal to the business to continue. This is exacerbated by monopolistic trends in the world where only one company is providing a good or service and buying out the competition when it raises to control choice. What we end up with is similar from the consumer's viewpoint to late Communism in regards to choice and quality. In the end Capitalism didn't win. It just lost last.

godelski15• 1 year ago

  > shipping features and fixes faster than ever before

Meanwhile Apple duplicated my gf's contract, creating duplicate birthdays on my calendar. It couldn't find duplicates despite matching name, nickname, phone number, birthdays, and that both contacts were associated with her Apple account. I manually merged and ended up with 3 copies of her birthday in my calendar...

Seriously, this shit can be solved with a regex...

The number of issues like these I see is growing exponentially, not decreasing. I don't think it's AI though, because it started before that. I think these companies are just overfitting whatever silly metrics they have decided are best

econ2• 1 year ago

The days of hating on idea men seem over.

I don't get it either. You hire someone in the hope for ROI. Some things work some kinda don't. Now people will be n times more productive therefore you should hire fewer people??

That would mean you have no ideas. It says nothing about the potential.

bawolff8• 1 year ago

Reistically its because layoffs have a high reputational cost. AI provides an excuse that lets companies do lay offs without suffering the reputation hit. In essence AI hype makes layoffs cheaper.

Doesnt really matter if AI actually works or not.

zelphirkalt3• 1 year ago

I would dispute, that there is no reputation cost, when you replace human work with LLMs.

strangattractor• 1 year ago

Most significant technology takes almost a generation to be fully adopted. I think it is unlikely we are seeing the full effect of LLM's at the moment.

Content producers are blocking scrapers of their sites to prevent AI companies from using their content. I would not assume that AI is either inevitable or on a easy path to adoption. AI certainly isn't very useful if what it "knows" is out of date.

asadotzler4• 1 year ago

In 10 years with the same amount of money and time that's been pumped into AI, still a financial black hole, we had the entire broadband internet build out completed and the internet was responsible for adding a trillion dollars a year to the global economy.

autobodie4• 1 year ago

No, we would see profits increase, and we have been seeing profits increase.

throwaway20379• 1 year ago

Regarding the impact of LLMs on non-programming tasks, check out this one:

https://www.ft.com/content/4f20fbb9-a10f-4a08-9a13-efa1b55dd...

    > The bank [Goldman Sachs] now has 11,000 engineers among its 46,000 employees, according to [CEO David] Solomon, and is using AI to help draft public filing documents.

    > The work of drafting an S1 — the initial registration prospectus for an IPO — might have taken a six-person team two weeks to complete, but it can now be 95 per cent done by AI in minutes, said Solomon.

    > “The last 5 per cent now matters because the rest is now a commodity,” he said.

In my eyes, that is major. Junior ibankers are not cheap -- they make about 150K USD per year minimum (total comp).

fourside• 1 year ago

This is certainly interesting and I don’t want to readily dismiss it, but I sometimes question how reliable these CEO anecdotes are. There’s a lot of pressure to show Wallstreet that you’re at the forefront of the AI revolution. It doesn’t mean no company is achieving great results but that it’s hard to separate the real anecdotes from the hype.

asadotzler4• 1 year ago

Claims by companies with an interest in AI without supporting documentation are just that, claims, and probably more PR and marketing than anything.

noisy_boy3• 1 year ago

I mean that's such a text heavy area anyway. I am not an expert in filing S1 but won't a lot of it be more or less boilerplate + customisations specific to the offering? Any reasonably advanced model should be able to take you a good chunk of the way. Then iterate with a verifier type model + a few people to review; even with iterations that should definitely shorten the overall time. It seems like such a perfect use case for an LLM - what am I missing that is hidden in the scepticism of the sibling comments?

wiseowise3• 1 year ago

I will never understand this argument. If you have a super tool, that can magically double your output, why would you suddenly double your output publicly? So that you now work twice essentially for the same money? You use it to work less, your output stays static or marginally improves - that’s smart play.

Note: I’m talking about your run of the mill SE waggie work, not startups where your food is based on your output.

conradkay• 1 year ago

That only works if you're one of very few people with the tool. Otherwise the rest of your team is now 2x as productive as you.

casualscience3• 1 year ago

In big companies, this is a bit slower due to the need to migrate entrenched systems and org charts into newer workflows, but I think you are seeing more productivity there too. Where this is much more obvious is in indie games and software where small agile teams can adopt new ways of working quickly...

E.g. look at the indie games count on steam by year: https://steamdb.info/stats/releases/?tagid=492

bojan• 1 year ago

The number of critically acclaimed games remains the same though. So for now we're getting quantity, but not the quality.

kaibee• 1 year ago

Has the amount of 95%+ reviews games-released increased though? And how much of that is due to the pandemic? Its anecdotal, but the game-dev discord I'm in has had a decent reduction in # of regulars since the tail end of the pandemic 24-25. And ironically, I was one of them until recently. I think people actually just had more time.

hansmayer14• 1 year ago

Well, it sort of evens out. You see the developers are pushed to use the AI to generate a lot of LoC-Slop, but then they have to fix all the bugs, security issues and hallucinated packages that were thrown in by the magic-machines. But at least some deluded MBA can BS about being "AI-first".

acrooks2• 1 year ago

I wonder if some of this output will take a while to be visible en masse.

For example, I founded a SaaS company late last year which has been growing very quickly. We are track to pass $1M ARR before the company's first birthday. We are fully bootstrapped, 100% founder owned. There are 2 of us. And we feel confident we could keep up this pace of growth for quite a while without hiring or taking capital. (Of course, there's an argument that we could accelerate our growth rate with more cash/human resources)

Early in my career, at different companies, we often solved capacity problems by hiring. But my cofounder and I have been able to turn to AI to help with this, and we keep finding double digit percentage productivity improvements without investing much upfront time. I don't think this would have been remotely possible when I started my career, or even just a few years ago when AI hadn't really started to take off.

So my theory as to why it doesn't appear to be "painfully obvious": you've never heard of most of the businesses getting the most value out of this technology, because they're all too small. On average, the companies we know about are large. It's very difficult for them to reinvent themselves on a dime to adapt to new technology - it takes a long time to steer a ship - so it will take a while. But small businesses like mine can change how we work today and realize the results tomorrow.

mixmastamyk2• 1 year ago

Curious, if you don’t mind mentioning what AIs you’re using (besides the obvious Claude, etc) and what for to augment your reach?

AndrewKemendo• 1 year ago

This is exactly how it’s going down

Companies that needed to hire 10 people to grow, only need to hire 9 now

In less than 5 years that’s going to be 7 or 6 people

I’m doing more with 5 engineers than I was able to do with 15 just 10 years ago

Part of that is libraries etc have matured too but we’ve reached the point from a developer perspective that you don’t need to build new technologies, you just need to put what exists together in new ways

All the parts exist for any technology to be built, it’s about composition and distribution at this point

aitchnyu• 1 year ago

Do existing teams (and ossified office politics) benefit from n-times faster devs? I witnessed (implied) Gantt charts so shaped that shrinking dev activities won't shrink the chart.

diego_moita• 1 year ago

That is a smart question.

In 1987 the economist Robert Solow said "You can see the computer age everywhere but in the productivity statistics".

We should remark he said this long before the internet, web and mobile, so probably the remark needs an update.

However, I think it cuts through the salesmen hype. Anytime we see these kinds of claims we should reply "show me the numbers". I'll wait until economists make these big claims, will not trust CEOs and salesmen.

marcosdumay3• 1 year ago

so probably the remark needs an update

Only if you want to add "internet, web, and mobile" before "age". Otherwise it doesn't need any change.

But that phrase is about the productivity statistics, not about computers or actual productivity.

zkry• 1 year ago

I find that this is on point. I've seen a lot of charts on the AI-hype side of things showing exponential growth of AI agent fleets being used for software development (starting in 2026 of course). Take this article for example: https://sourcegraph.com/blog/revenge-of-the-junior-developer

Ok, so by 2027 we should be having fleets of autonomous AI agents swarming around every bug report and solving it x times faster than a human. Cool, so I guess by 2028 buggy software will be a thing of the past (for those companies that fully adopt AI of course). I'm so excited for a future where IT projects stop going overtime and overbudget and deliver more value than expected. Can you blame us for thinking this is too good to be true?

vharish• 1 year ago

Overall, the amount of code that's being deployed to production has definitely increased.

hombre_fatal3• 1 year ago

This is like asking if tariffs are so bad, why don't you notice large price swings in your local grocer right now?

In complex systems, you can't necessarily perceive the result of large internal changes, especially not with the tiny amount of vibes sampling you're basing this on.

You really don't have the pulse on how fast the average company is shipping new code changes, and I don't see why you think you would know that. Shipping new public end-use features isn't even a good signal, it's a downstream product and a small fraction of software written.

It's like thinking you are picking up a vibe related to changes in how many immigrants are coming into the country month to month when you walk around the mall.

jayd164• 1 year ago

Maybe not a great analogy. The market reacted instantly and you can see prices fluctuate almost as fast as tariff policy.

mNovak• 1 year ago

I mean, if a mega corp like Google or Amazon had plus/minus 10% of their headcount, as a lay observer I don't think I'd really be able to detect the difference in output either.

That doesn't mean it isn't a real productivity gain, but it might be spread across enough domains (bugs, features, internal tools, experiments) to not be immediately or "painfully obvious".

It'll probably get more obvious if we start to see uniquely productive small teams seeing success. A sort of "vibe-code wonder".

antithesizer3• 1 year ago

Before enterprise AI systems are allowed to spread their wings, first they need to support existing processes. Once they're able to generate the same customer-facing results relatively autonomously, then they'll have the opportunity to improve those results. So the first place to look for their impact is, I'd wager, cost-cutting. So watch those quarterly earnings reports.

0xbadcafebee• 1 year ago

Productivity results in increased profit, not necessarily output. They don't need to innovate, make new products, or improve things. They just need to make their shit cheaper so their profit margin is higher. If you can just keep churning out more money, there is no need to improve anything.

marcosdumay3• 1 year ago

I don't think extra productivity in software development ever reflected in established companies building things faster.

The more likely scenario is that if those tools make developer so much more productive, we would see a large surge in new companies, with 1 to 3 developers creating things that were deemed too hard for them to do.

But it's still possible that we didn't give people enough time yet.

whstl• 1 year ago

Any boost of productivity in the coding part is quickly absorbed by other inefficiencies in the software-making process, unfortunately.

AI also helps immensely in creating those other inefficiencies.

klabb3• 1 year ago

Intuitively I agree. In the long run, we’ll know better. But for now, nobody truly knows what the new equilibrium is.

That said: it’s one type of work that is getting dramatically cheaper. The debate is about the scope and quality of that labor, not whether it’s cheap or fast (it is). But if anything negative (errors, faults) compound, and the correction can NOT be done with the same tools, then you must still have humans triage errors. In my experience, bad code can already have negative value (it costs more to fix than rewrite).

In the medium term, the actual scope and ability for different tasks will remain unknown. It takes a lot of time to gather the experience to tell if something was a bad idea – just look at the graveyard of design patterns, languages and software practices. Many of them enjoyed the spotlight for a decade before the fallout hit.

Anyway, while the abilities are unknown, AI will be used everywhere for everything – which is only wise if it’s truly better at every general task – despite every available data about it shows vastly different ability in different domains/problem types. Many of those things will be both (a) worse than humans and (b) expensive to reverse, with compounding effects.

The funny thing is I have already seen enthusiasts basically acknowledging this but explaining that those compounding issues (think tech debt) is the right choice now because better AI will fix those issues in the future. To me, this feels like the early formations of religion (not metaphorically even). And I have a feeling that the goalpost moving from both sides will lead to an unfalsifiability deadlock in the debate.

JackYoustra• 1 year ago

my favorite solow line "You can see the computer age everywhere but in the productivity statistics.": https://en.wikipedia.org/wiki/Productivity_paradox

This eventually changed. Companies do figure out how to use tech, it just takes a while.

arthurcolle• 1 year ago

https://github.com/arthurcolle/mlx.erl

simonsarris3• 1 year ago

I think the real white collar bloodbath is that the end of ZIRP was the end of infinite software job postings, and the start of layoffs. I think its easy to now point to AI, but it seems like a canard for the huge thing that already happened.

just look at this:

https://fred.stlouisfed.org/graph/?g=1JmOr

In terms of magnitude the effect of this is just enormous and still being felt, and never recovered to pre-2020 levels. It may never. (Pre-pandemic job postings indexed to 100, its at 61 for software)

Maybe AI is having an effect on IT jobs though, look at the unique inflection near the start of 2025: https://fred.stlouisfed.org/graph/?g=1JmOv

For another point of comparison, construction and nursing job postings are higher than they were pre-pandemic (about 120 and 116 respectively, where pre-pandemic was indexed to 100. Banking jobs still hover around 100.)

I feel like this is almost going to become lost history because the AI hype is so self-insistent. People a decade from now will think Elon slashed Twitter's employee count by 90% because of some AI initiative, and not because he simply thought he could run a lot leaner. We're on year 3-4 of a lot of other companies wondering the same thing. Maybe AI will play into that eventually. But so far companies have needed no such crutch for reducing headcount.

leflambeur4• 1 year ago

It's simply the old Capital vs Labor struggle. CEOs and VCs all sing in the same choir, and for the past 3 years the tune is "be leaner".

p.s.: I'm a big fan of yours on Twitter.

godelski15• 1 year ago

  > the tune is "be leaner".

Seems like they're happy to start cutting limbs to lose weight. It's hard to keep cutting fat if you've been aggressively cutting fat for so long. If the last CEO did their job there shouldn't be much fat left

saubeidl• 1 year ago

Except Labor in Tech is unique in that it has zero class consciousness and often actively roots for their exploiters.

If we were to unionize, we could force this machine to a halt and shift the balance of power back in our favor.

But we don't, because many of us have been brainwashed to believe we're on the same side as the ones trying to squeeze us.

digitcatphd• 1 year ago

As of now yes. But we are still in day 0.1 of GenAI. Do you think this will be the case when o3 models are 10x better and 100x cheaper? There will be a turning point but it’s not happened yet.

nradov4• 1 year ago

10× better by what metric? Progress on LLMs has been amazing but already appears to be slowing down.

nothercastle• 1 year ago

I think they will be 10-100x cheaper id be really surprised if we even doubled the quality though

makeitdouble• 1 year ago

How does it work if they get 10x better in 10 years ? Everything else will have already moved on and the actual technology shift will come from elsewhere.

Basically, what if GenAI is the Minitel and what we want is the internet.

directevolve2• 1 year ago

We’re already heading toward the sigmoid plateau. The GPT 3 to 4 shift was massive. Nothing since had touched that. I could easily go back to the models I was using 1-2 years ago with little impact on my work.

I don’t use RAG, and have no doubt the infrastructure for integrating AI into a large codebase has improved. But the base model powering the whole operation seems stuck.

apwell232• 1 year ago

Do you think this will be the case when o3 models are 10x better and 100x cheaper?

why don't you bring it up then.

There will be a turning point but it’s not happened yet.

do you know something that rest of us don't ?

ricardobayes• 1 year ago

Frankly, we don't know. That "turning point" that seemed so close for many tech, never came for some of them. Think 3D-printing that was supposed to take over manufacturing. Or self-driving, that is "just around the corner" for a decade now. And still is probably a decade away. Only time will tell if GenAI/LLMs are color TV or 3D TV.

threeseed4• 1 year ago

How are we in 0.1 of GenAI ? It's been developed for nearly a decade now.

And each successive model that has been released has done nothing to fundamentally change the use cases that the technology can be applied to i.e. those which are tolerant of a large percentage of incoherent mistakes. Which isn't all that many.

So you can keep your 10x better and 100x cheaper models because they are of limited usefulness let alone being a turning point for anything.

godelski15• 1 year ago

Yet we're what? 5 years into "AI will replace programmers in 6 months"?

10 years into "we'll have self driving cars next year"

We're 10 years into "it's just completely obvious that within 5 years deep learning is going to replace radiologists"

Moravec's paradox strikes again and again. But this time it's different and it's completely obvious now, right?

croes• 1 year ago

If not when.

solumunus• 1 year ago

I use LLM’s daily and live them but at the current rate of progress it’s just not really something worth worrying about. Those that are hysterical about AI seem to think LLM’s are getting exponentially better when in fact diminishing returns are hitting hard. Could some new innovation change that? It’s possible but it’s not inevitable or at least not necessarily imminent.

AvAn122• 1 year ago

Remember when RPA was going to replace everyone?

johnnyanmac3• 1 year ago

There's a lot of "when" people are betting on, and not a lot of action to back it. If "when" is 20 years, then I still got plenty career ahead of me before I need to worry about that.

xnx2• 1 year ago

5 years into "AI will replace programmers in 6 months"?

Programmers that don't use AI will get replaced by those that do. (no just by mandate, but by performance)

10 years into "we'll have self driving cars next year"

They're here now. Waymo does 250K paid rides/week.

bawolff8• 1 year ago

Honestly, if anything i think AI is going to reverse the trend. Someone is going to have to be hired to clean up after them.

tempodox• 1 year ago

If anyone will actually bother with cleaning up.

xkcd19635• 1 year ago

Thats the impression I got. Things overall get just worse in quality because people rely too much on low wages and copy pasting LLM answers

notepad0x90• 1 year ago

I think they said that about outsourcing software dev jobs. The reality is somewhere in the middle. extreme cases will need cleanup but overall it's here to stay, maybe with more babysitting.

jameslk• 1 year ago

Keynes suggested that by 2030, we’d be working 15 hour workweeks, with the rest of the time used for leisure. Instead, we chose consumption, and helicopter money gave us bullshit jobs so we could keep buying more bullshit. This is fairly evident by the fact when the helicopter money runs out, all the bullshit jobs get cut.

AI may give us more efficiency, but it will be filled with more bullshit jobs and consumption, not more leisure.

pmlnr• 1 year ago

Keynes was talking about work in every sense,including house chore. We're well below 15 hours of house chores by now, so that part became true.

autobodie4• 1 year ago

Keynes lived in a time when the working class was organized and exerting its power over its destiny.

We live in a time that the working class is unbelievably brainwashed and manipulated.

SarahC_• 1 year ago

"Bullshit jobs" are the rubbish required to keep the paperwork tidy, assessed and filed. No company pays someone to do -nothing-.

AI isn't going to generate those jobs, it's going to automate them.

ALL our bullshit jobs are going away, and those people will be unemployed.

davedx• 1 year ago

Some countries are still trending in that direction:

https://www.theguardian.com/commentisfree/2024/nov/21/icelan...

Policy matters

tim33316• 1 year ago

I think something Keynes got wrong there and much AI job discussion ignores is people like working, subject to the job being fun. Look at the richest people with no need to work - Musk, Buffett etc. Still working away, often well past retirement age with no need for the money. Keynes himself, wealth and probably with tenure working away on his theories. In the UK you can quite easily do nothing by going on disability allowance and doing nothing and many do but they are not happy.

There can be a certain snobbishness with academics where they are like of course I enjoy working away on my theories of employment but the unwashed masses do crap jobs where they'd rather sit on their arses watching reality TV. But it isn't really like that. Usually.

seydor• 1 year ago

Most of the people are leisuring af work (for keynes era standards) and also getting paid for it

ccorcos13• 1 year ago

If you work 15 hours/week then presumably someone who chose to work 45 hours/week would make 3x more money.

This creates supply-demand pressure for goods and services. Anything with limited supply such as living in the nice part of town will price out anyone working 15 hours/week.

And so society finds an equilibrium…

JumpCrisscross5• 1 year ago

Keynes suggested that by 2030, we’d be working 15 hour workweeks

Most people with a modest retirement account could retire in their forties to working 15-hour workweeks somewhere in rural America.

gosub1002• 1 year ago

Instead, we chose consumption

instead, corporations chose to consume us

Slow_Hand2• 1 year ago

Not to undercut your point - because you’re largely correct - but this is my reality. I have a decent-paying job in which I work roughly 15 hrs a week. Sometimes more when work scales up.

That said, I’m not what you’d call a high-earning person (I earn < 100k) I simply live within my means and do my best to curb lifestyle creep. In this way, Keynes’ vision is a reality, but it’s a mindset and we also have to know when enough wealth is enough.

raincom• 1 year ago

Now one has to work 60 hours to afford housing(rent/mortgage) and insurance (health, home, automotive). Yes, food is cheap if one can cook.

WorkerBee28474• 1 year ago

Keynes suggested that by 2030, we’d be working 15 hour workweeks

Yeah, I'd say I get up to 15 hours of work done in a 40 hour workweek.

laughing_man2• 1 year ago

It's more likely 15% of the workforce will have jobs. They'll be working eighty hour weeks and making just enough to keep them from leaving.

latentsea• 1 year ago

It's still not 2030 yet. It could still happen.

runeks2• 1 year ago

Keynes also convinced us that high unemployment and high inflation couldn't happen at the same time. This was proven wrong in the early 1970s.

onlyrealcuzzo2• 1 year ago

Keynes suggested that by 2030, we’d be working 15 hour workweeks, with the rest of the time used for leisure.

I suspect he didn't factor in how may people would be retired and on entitlements.

We're not SUPER far from that now, when you factor in how much more time off the average person has now, how much larger of percentage of the population is retired, and how much of a percentage is on entitlements.

The distribution is just very unequal.

I.E. if you're the median worker, you've probably seen almost no benefit, but if you're old or on entitlements, you've seen a lot of benefits.

csomar3• 1 year ago

Elon Musk experiment is the worst anchor that can be used for comparison since the dude destabilized Twitter (re-branding, random layoffs, etc...). I'd be more interested in companies that went leaner but did it in a sane manner. The Internet user base grew between 2022 and now but Twitter might have lost users in that time period and certainly didn't make any new innovations beyond trying to charge its users more and confusing them.

e403• 1 year ago

Also section 174’s amortization of software development had a big role.

jki275• 1 year ago

That's about to get repealed it looks like.

Lu20257• 1 year ago

I agree, R&D change is what triggered 2022 tech layoffs. Coders used to be free, all this play with Metaverse and such was on public dime. As soon as a company had to spend real money, it all came crashing down.

lozenge• 1 year ago

Macroeconomic policy always changes, recessions come and go, but it's not a permanent change in the way e-commerce or AI is.

gmerc• 1 year ago

The flaw with the Zirp narrative that companies managed to raise more money than ever before the moment they had a somewhat believable narrative instead of the crypto/web3/metaverse nonsense.

xorcist3• 1 year ago

because he simply thought he could run a lot leaner

Because he suddenly had to pay interest on that gigantic loan he (and his business associates) took to buy Twitter.

It may not be the only reason for everything that happened, but it sure is simple and has some very good explanatory powers.

huntertwo3• 1 year ago

Other companies have different reasons to cut costs, but the incentive is still there.

perrygeo• 1 year ago

Such an important point, I've seen and suspected the end of ZIRP being a much much greater influence on white collar work than we suspect. AI is going to take all the negative press but the flow of capital is ultimately what determines how the business works, which determines what software gets built. Conway's law 101. The white collar bloodbath is more of a haircut to shed waste accumulated during the excesses of ZIRP.

steveBK1233• 1 year ago

ZIRP and then the final gasp of COVID bubble over hiring.

At least in my professional circles the number of late 2020-mid 2022 job switchers was immense. Like 10 years of switches condensed into 18-24 months.

Further lot of experiences and anecdotes talking to people who saw their company/org/team double or triple in size when comparing back to 2019.

Despite some waves of mag7 layoffs we are still I think digesting what was essentially an overhiring bubble.

steve_adams_86• 1 year ago

Is it negative press for AI, or is it convincing some investors that it’s actually causing a tectonic shift in the workforce and economy? It could be positive in some sense. Though ultimately negative, because the outcomes are unlikely to reflect a continuation of the perceived impact or imaginary progress of the technology.

hoosieree3• 1 year ago

AI also happens to be a perfect scapegoat: CEOs who over-hired get to shift the blame to this faceless boogeyman, and (bonus!) new hires are more desperate/willing to accept worse compensation.

fallingknife• 1 year ago

ZIRP had little to do with it. Tech is less levered than any other major industry. What happened is that growth expectations for large tech companies were way out of line with reality and finally came back down to earth when the market finally realized that the big tech cos are actually mature profitable companies and not just big startups. The fact that this happened at the same time ZIRP ended is a coincidence.

Mistletoe• 1 year ago

This is so cool. Had no idea FRED had data like this. They have everything.

cyanydeez• 1 year ago

give trump a few more years, and that probably will change.

hombre_fatal3• 1 year ago

Why would you interpret data cut off at 2020 so that you're just looking at a covid phenomenon? The buttons don't seem to do anything on that site, but why not consider 2010-2025?

That said, the vibe has definitely shifted. I started working in software in uni ~2009 and every job I've had, I'd applied for <10 positions and got a couple offers. Now, I barely get responses despite 10x the skills and experience I had back then.

Though I don't think AI has anything to do with it, probably more the explosion of cheap software labor on the global market, and you have to compete with the whole world for a job in your own city.

Kinda feels like some major part of the gravy train is up.

lbotos• 1 year ago

It looks like that specific graph only starts in 2020...

oblio5• 1 year ago

People a decade from now will think Elon slashed Twitter's employee count by 90% because of some AI initiative, and not because he simply thought he could run a lot leaner.

That part is so overblown. Twitter was still trying to hit moonshots. X is basically in "keep the lights on" mode as Musk doesn't need more. Yeah, if Google decides it doesn't want to grow anymore, it can probably cut it's workforce by 90%. And it will be as irrelevant as IBM in maximum 10 years.

aikinai• 1 year ago

What moonshots has Twitter gone for in the last decade? Feature velocity is also higher since the acquisition.

leptons2• 1 year ago

What's happening now is similar to what happened during the 2000's "dot-com bubble burst". Having barely survived that time, I saw this one coming and people told me I was crazy when I told them to hold on to their jobs and quit job-hopping, because the job-hopper is very often the first one to get laid off.

In 2000 I was moved cities and I had a job lined-up at a company that was run by my friends, I had about 15 good friends working at the company including the CEO, and I was guaranteed the job in software development at the company. The interview was supposed to be just a formality. So I moved, and went in to see the CEO, and he told me he could not hire me, the funding was cut and there was a hiring freeze. I was devastated. Now what? Well I had to freelance and live on whatever I could scrape together, which was a few hundred bucks a month, if I was lucky. Fortunately the place I moved into was a big house with my friends who worked at said company, and since my rent was so low at the time, they covered me for a couple of years. I did eventually get some freelance work from the company, but things did not really recover until about 2004 when I finally got a full-time programming job, after 4 very difficult years.

So many tech companies over-hired during covid, there was a gigantic bubble happening with FAANG and every other tech company at the time. The crash in tech jobs was inevitable.

I feel bad for people who got left out in the cold this time, I know what they are going through.

yellow_lead• 1 year ago

Those are some great friends. Aside from job hoppers, I noticed there are a lot of company loyalists getting canned too though (i.e worked at MSFT 10 years)

intalentive• 1 year ago

Yes. Tech is clearly a beneficiary of the Cantillon Effect.

rglover3• 1 year ago

IMO this is dead on. AI is a hell of a scapegoat for companies that want to save face and pretend that their success wasn't because of cheap money being pumped into them. And in a world addicted to status games, that's a gift from the heavens.

esperent2• 1 year ago

ZIRP is an American thing? In that case maybe we could try comparisons with the job markets in other developed Western countries that didn't have this policy. If it was because of ZIRP, then their job markets should show clearly different patterns.

jt2190• 1 year ago

It’s a shame that this is the top comment because it’s backward looking (“here’s why white-collar workers lost their jobs in the last year”) instead of looking forward and noticing that even if interest rates are reduced back to zero these jobs will not be performed by humans ever again. THAT is the message here. These workers need to retrain and move on.

SR2Z• 1 year ago

even if interest rates are reduced back to zero these jobs will not be performed by humans ever again

It's not like companies laid off whole functions. These jobs will continue to be performed by humans - ZIRP just changes the number of humans and how much they get paid.

These workers need to retrain and move on.

They only need to "retrain" insofar as they keep up with the current standards and practices. Software engineers are not going anywhere.

niuzeta• 1 year ago

FRED continues to amaze me with the kind of data they have availab.e

brfox• 1 year ago

That's from Indeed. And, Indeed has fewer job postings overall [https://fred.stlouisfed.org/series/IHLIDXUS]. Should we normalize the software jobs with the total number of Indeed postings? Is Indeed getting less popular or more popular over this time period? Data is complicated

treyd• 1 year ago

That inflection point seems to more specifically start at the day of the new administration's inauguration.

wonderwonder2• 1 year ago

Saw something similar the other day. X was awash with stories that IBM was laying off several thousand people in their HR dept. being let go due to Ai. Then over the course of the day the story shifted to IBM was outsourcing them all to India. Was a very interesting transition, seemed intentional.

Lu20257• 1 year ago

IBM seemed to outsource recruiting to Indian firms too and it's awful. The accounts who contact me on LinkedIn are grossly unprofessional and downright nasty.

alfiedotwtf2• 1 year ago

Bingo!

Lu20257• 1 year ago

unique inflection near the start of 2025

I wonder what happened in January 2025...

jalapenos• 1 year ago

Always disheartening how much people forget and tolerate the underlying deliberate human absurdity that created these events.

Almost no one has seen a world where the price of money wasn't centrally planned, a committee of experts deciding it based on gut feel like they did in command economies like the Soviet union.

And then thousands of people's lives are disrupted as the interest rate swings wildly due purely to government action (corona lockdowns and fed zirp response), and it all somehow just ends up people talking about AI instead.

The true wrongdoers get absolutely no consequences, and we all just carry on like there's no problem. Often because our taxes go to paying hordes of academics and economists to produce layers and layers of sophisticated propaganda that of course this system is the best one.

Absurd and shitty world.

1vuio0pswjnm72• 1 year ago

ZIRP jobs, n., jobs the compensation for which is derived from zero interest loans, often in the form of venture capital, instead of reserves, profits or other sources

"When interest rates return to normal levels, the ZIRP jobs will disappear." -- Wall Street analyst

fennecfoxy• 1 year ago

Inflation & mismanagement.

randomname4325• 1 year ago

Only way to know for sure you're safe from replacement is if your job is a necessary part of something generating revenue and your not easily replaceable. Otherwise you should assume the company won't hesitate to replace you. It's just business.

Tokkemon• 1 year ago

Yeah I thought that too. Then they laid me off anyway.

snackernews• 1 year ago

Anyone who thinks an executive considers them necessary or irreplaceable in the current environment is fooling themselves.

sailingparrot3• 1 year ago

if your job is a necessary part of something generating revenue and your not easily replaceable.

First part of this statement is clearly false. People on the phone in a tech support company are very much necessary to generate revenue, people tending to field were very much necessary to extract the value of the fields. Draftsmen before CAD were absolutely necessary etc.

Yet technology replaced them, or is in the process of doing so.

So then, your statement simplifies to “if you want to be safe for replacement have a job that’s hard to replace” which isn’t very useful anymore.

econ2• 1 year ago

There use to be a cookie factory here that had up to 12 people sitting there all day doing nothing. If the machines broke down it really took all of them to clean up. This pattern will be rediscovered.

topherPedersen• 1 year ago

I could be wrong, but I think us software developers are going to become even more powerful, in demand, and valuable.

ghm2180• 1 year ago

I wonder when the investors and investors in the early printing press or steam engine or excel spreadsheet was invented did they think of the ways — soul crushing homework(books), rapid and cruel colonization(steam engines and trains), innovative project management(excel) — there tech would be used?

The demand for these products was not where it was intended at the time probably. Perhaps the answer to its biggest effect lies in how it will free up human potential and time.

If AI can do that — and that is a big if — then how and what would you do with that time? Well ofc, more activity, different ways to spend time, implying new kinds of jobs.

HarHarVeryFunny5• 1 year ago

The trouble with looking at past examples of new tech and automation is that those were all verticals - the displaced worker could move to a different, maybe newly created, work area left intact by the change.

Where AI will be different (when we get there - LLMs are not AGI) is that it is a general human-replacement technology meaning there will be no place to run ... They may change the job landscape, but the new jobs (e.g. supervising AIs) will ALSO be done by AI.

I don't buy this "AGI by 2027" timeline though - LLMs and LLM-based agents are just missing so many basic capabilities compared to a human (e.g. ability to learn continually and incrementally). It seems that RL, test-time compute (cf tree search) and agentic application, have given a temporary second wind to LLMs which were otherwise topping out in terms of capability, but IMO we are already seeing the limits of this too - superhuman math and coding ability (on smaller scope tasks) do not translate into GENERAL intelligence since they are not based on general mechanism - they are based on vertical pre-training in these (atypical in terms of general use case) areas where there is a clean reward signal for RL to work well.

It seems that this crazy "we're responsibly warning you that we're going to destroy the job market!" spiel is perhaps because these CEOs realize there is a limited window of opportunity here to try to get widespread AI adoption (and/or more investment) before the limitations become more obvious. Maybe they are just looking for an exit, or perhaps they are hoping that AI adoption will be sticky even if it proves to be a lot less capable that what they are promising it will be.

infinitebit2• 1 year ago

I am SO thankful to see a news outlet take what tech CEOs say with a grain of salt re: AI. I feel like so many have just been breathlessly repeating anything they say without even an acknowledgement that there might be, you know, some incentive for them to stretch the truth.

(ftr i’m not even taking a side re: will AI take all the jobs. even if they do, the reporting on this subject by MSM has been abysmal)

infinitebit2• 1 year ago

So glad to see a MSM outlet take the words of an AI ceo with even a single grain of salt. I’ve been really disappointed with the way so many publications have just been breathlessly repeating what is essentially a sales pitch.

(ftr i’m not even taking a side re: is AI going to take all the jobs. regardless of what happens the fact remains that the reporting has been absolute sh*t on this. i guess “the singularity is here” gets more clicks than “sales person makes sales pitch”)

absurdo3• 1 year ago

HN does the same. We don’t really have a platform on the internet for good discussions so we mostly get regurgitated talking points and a lot of flags/downvotes if it’s deemed a serious enough issue (f.e. pandemic) that taking a contrary stance is strictly forbidden.

throwaway3141552• 1 year ago

Case in point ^

brokegrammer• 1 year ago

We don't need AI to wipe out entry-level office jobs. David Graeber wrote about this in Bullshit Jobs. But now that we have AI, it's a good excuse to wipe out those jobs for good, just like Elon did after he acquired Twitter. After that, we can blame AI for the deed.

JanisErdmanis• 1 year ago

The productivity gains in activities will be countered by the same gains in counter activities. Everything is going to become more sophisticated, but bullshit will remain.

kilroy123• 1 year ago

I've thought a lot about this. I think this is exactly what is happening. I've seen this first hand.

A lot of the BS jobs are being killed off. Do some non-bs jobs get burn up in the fire along the way, yes. But it's only the beginning.

rule2025• 1 year ago

The real "white-collar massacre" is not caused by AI, but you have no irreplaceable, or the value created by hiring you is not higher than using AI. Businesses will not hesitate to use AI, you can't say that companies are ruthless, but that's the pursuit of efficiency. Just as horse-drawn carriages were replaced by cars and coachmen lost their jobs, you can't say it's a problem with cars.

History is always strikingly similar, the AI revolution is the fifth industrial revolution, and it is wise to embrace AI and collaborate with AI as soon as possible.

HarHarVeryFunny5• 1 year ago

There's a popular saying, e.g. used by NVIDIA CEO Jensen Huang, that "AI won't replace you - a human using AI will replace you", which may be temporarily true while AI isn't very capable, but the AI CEOs are claiming AGI will be here in 2 years, and explicitly saying that it will be a "drop-in replacement remote worker". Obviously one of these is wrong - it's either just a tool to be learnt and used, or it is in fact a drop-in replacement for a human.

One can argue about the timeline and technology (maybe not LLM based), but it does seem that human-level AGI will be here relatively soon - next 10 or 20 years, perhaps, if not 2. When this does happen, history is unlikely to be a good predictor of what to expect... AGI may create new jobs as well as detstroy old ones, but what's different is that AGI will also be doing those new jobs! AGI isn't automating one industry, or creating a technology like computers that can help automate any industry - AGI is a technology that will replace the need for human workers in any capacity, starting with all jobs that can be conducted without a physical presence.

notyouraibot• 1 year ago

The hype around AI replacing software engineers is truly delusional. Yes they are very good at solving known problems, writing for loops and boilerplate code but introduce a little bit of complexity and creativity and it all fails. There have been countless tasks that I have given to AI, to which it simply concluded its not possible and suggested me to use several external libraries to get it done, after a little bit of manual digging, I was able to achieve that same task without any libraries and I'm not even a seasoned engineer.

jatora• 1 year ago

While I agree that the current 'bloodbath' narrative is all hype, I'm honestly confused by a lot of the sentiment i see on here towards AI. Namely the dismissal of continual improvement and the rampant whistling past the graveyard attitude of what is coming.

It is confusing because many of the dismissals come from programmers, who are unequivocally the prime beneficiaries of genAI capability as it stands.

I work as a marketing engineer at a ~1B company and the amount of gains I have been able to provide as an individual are absolutely multiplied by genAI.

One theory I have is that maybe it is a failing of prompt ability that is causing the doubt. Prompting, fundamentally, is querying vector space for a result - and there is a skill to it. There is a gross lack of tooling to assist in this which I attribute to a lack of awareness of this fact. The vast majority of genAI users dont have any sort of prompt library or methodology to speak of beyond a set of usual habits that work well for them.

Regardless, the common notion that AI has only marginally improved since GPT-4 is criminally naive. The notion that we have hit a wall has merit, of course, but you cannot ignore the fact that we just got accurate 1M context in a SOTA model with gemini 2.5pro. For free. Mere months ago. This is a leap. If you have not experienced that as a leap then you are using LLM's incorrectly.

You cannot sleep on context. Context (and proper utilization of it) is literally what shores up 90% of the deficiencies I see complained about.

AI forgets libraries and syntax? Load in the current syntax. Deep research it. AI keeps making mistakes? Inform it of those mistakes and keep those stored in your project for use in every prompt.

I consistently make 200k+ token queries of code and context and receive highly accurate results.

I build 10-20k loc tools in hours for fun. Are they production ready? No. Do they accomplish highly complex tasks for niche use cases? Yes.

The empowerment of the single developer who is good at manipulating AI AND an experienced dev/engineer is absolutely incredible.

Deep research alone has netted my company tens of millions in pipeline, and I just pretend it's me. Because that's the other part that maybe many aren't realizing - its right under your nose - constantly.

The efficiency gains in marketing are hilariously large. There are countless ways to avoid 'AI slop', and it involves, again, leveraging context and good research, and a good eye to steer things.

I post this mostly because I'm sad for all of the developers who have not experienced this. I see it as a failure of effort (based on some variant of emotional bias or arrogance), not a lack of skill or intellect. The writing on the wall is so crystal clear.

sfblah2• 1 year ago

You're right, of course. Most of this thread is some sort of weird motivated reasoning by people who are terrified of the reality of what lies ahead. I work at a top-10 tech company. We've stopped hiring junior talent, and it's 100% because of AI. I'm something like 2x more productive since using AI. We're now deploying agentic AI systems to further reduce headcount. The actual bloodbath will happen when there's any kind of financial pressure on the company (a recession).

Lu20257• 1 year ago

when there's any kind of financial pressure on the company (a recession)

When was the last time there was no financial pressure? I keep hearing how hard it is to be a small business owner for as long as I'm an adult and it's like a quarter century by now.

atleastoptimal2• 1 year ago

losing jobs is the biggest predictable hazard of AI but far from the biggest

however there seems to be a big disconnect on this site and others

If you believe AGI is possible and that AI can be smarter than humans in all tasks, naturally you can imagine many outcomes far more substantial than job loss.

However many people don’t believe AGI is possible, thus will never consider those possibilities

I fear many will deny the probability that AGI could be achieved in the near future, thus leaving themselves and others unprepared for the consequences. There are so many potential bad outcomes that could be avoided merely if more smart people realized the possibility of AGI and ASI, and would thus rationally devote their cognitive abilities to ensuring that the potential emergence of smarter than human intelligences goes well.

theawakened• 1 year ago

I've said this before and I'll say it again: The idea that 'AI' will EVER take over any programmers job is ridiculous. These idiots think they are going to create AGI, it's never going to happen, not with this race of people. There is far too much ignorance in humanity. AI will never be able to be any better than it's source, humanity. It's a soon-to-be realization for these billionaire talking heads. Nothing can rise higher than it's source. Even if they cover every square foot of land with data centers, it'll never work like they expect it to. The AI bubble will burst so hard the entire world will quake. I give it 5 years max.

hansmayer14• 1 year ago

It's so good to see the non-expert types are finally starting to see the whole hype for what it really is -> the long tail of last 20 years of incremental ML development, and not some revolutionary tech. We did not need to have this much hype around web 1.0 which was immediately adopted due to being obviously, well, revolutionary.

madaxe_again6• 1 year ago

Yes, the dot-com bubble never happened.

hansmayer14• 1 year ago

We're talking technology adoption here, no need to sidetrack (although I would argue, the bubble of technology sector burning 200B to produce 10B in revenue will be so much more painful). Back in the day everyone was running to do something on the web, commercial or not, be it out of enthusiasm, greed, noble aspirations, art or even criminal (warez, anyone?). I won't even mention the iPhone moment here. Compare that to five years into GenAI hype, where is the massive adoption and thousands of applications or at least a single important breakthrough? Where is the AGI some of these "leaders" have been promising to arrive in 2025? Chumming along with Tesla's robotaxis?

K0balt2• 1 year ago

The “bloodbath” will be slow but is quite likely to be significant.

AI / GP robotic labor will not penetrate the market so much in existing companies, which will have huge inertial buffers, but more in new companies that arise in specific segments where the technology proves most useful.

The layoffs will come not as companies replace workers with AI, but as AI companies displace non-AI companies in the market, followed by panicked restructuring and layoffs in those companies as they try to react, probably mostly unsuccessfully.

Existing companies don’t have the luxury of buying market share with investor money, they have to make a profit. A tech darling AI startup powered by unicorn farts and inference can burn through billions of SoftBank money buying market share.

xpe2• 1 year ago

I find this plausible. Is the data starting to show this?

K0balt2• 1 year ago

There might be some, but I think it’s still early.

For the moment, AI is enabling a bunch of stuff that was too expensive or time consuming to do before (flooding the commons with shiny garbage and pedantic text to drive “engagement”.

Despite the hype, It’s going to be 2-3 years before AI application really fall into stride, and 3-7 before general purpose robotics really get up to speed.

veunes• 1 year ago

If the product can't speak for itself, scare people into believing it will soon

jona777than• 1 year ago

There will likely be more jobs because of AI. With more “knowledge”, comes more responsibility. Spam folders only exist because of automated emails. That classification process is more work. We may find there are more needs to meet as AI advances, not less.

The fallacy is in the statement “AI will replace jobs.” This shirks responsibility, which immediately diminishes credibility. If jobs are replaced or removed, that’s a choice we as humans have made, for better or worse.

givemeethekeys3• 1 year ago

Many people are unable to find jobs because they are too old.

Even older people prefer to hire younger people.

throwaway3141552• 1 year ago

Okay? What does that have to do with anything?

givemeethekeys3• 1 year ago

Losing one’s job is only half of the story. Finding another one when you’re getting old in the AI age is more difficult than before AI.

givemeethekeys3• 1 year ago

There is an AI bloodbath that is adding to supply of labor in all low hanging fields that aren’t yet being decimated by AI.

franczesko• 1 year ago

AI bubble burst will come first.

Lu20257• 1 year ago

don't think that the white collar layoffs of the last 3-4 years are due to AI. The tech layoffs of 2022 are explained in part by the impact of 2017 tax reform. Before 2022, the research and development expenses could be written off taxes as a tax credit in the same year. It's a dollar to dollar reduction of tax liability. Tech companies classify a lot of their work as R&D. So those overpaid Facebook coders are essentially public charges! Somebody up thread said that programmers are disproportionately highly compensated in the US. They are, because it's not companies who pay for them, it's taxpayers, indirectly. Starting 2022, the deal became a bit less sweet. R&D expenses had to be amortized over 5 years. What happened next is a collusion in response to Great Resignation. Several large tech companies conspired to have layoffs at the same time as a salary compression move. The AI statements are mostly a scare tactic to put pressure on employees. For some industries and applications AI is revolutionary, but for coding it's good at autocomplete and not much else.

nova220332• 1 year ago

Does anyone have any experience with using AI tools on a massive legacy code base?

HenryBemis• 1 year ago

To be clear, Amodei didn’t cite any research or evidence for that 50%

This reminds me the "Walter White" meme "I am the documentation". When the CEO of a company that makes LLM says something like that, "I perk up and listen" (to quote the article).

When a doctor says "water in my village is bad quality, it gives diarrhea to 30% of the villagers", I don't need a fancy study from some university. The doctor "is the documentation". So if the Anthropic/ChatGPT/LLaMa/etc. (mixing companies and products, it's ok though) say that "so-and-so", they see the integrations, enhancements, compliments, companies ordering _more_ subscriptions, etc.

In my current company (high volume, low profit margin) they told us "go all in on AI". They see that (e.g. with Notion-like-tools) if you enable the "AI", that thing can save _a lot_ of time on "Confluence-like" tasks. So, paying $20-$30-$40 per person, per month, and that thing improving the productivity/output of an FTE by 20%-30% is a massive win.

So yes, we keep the ones we got (because mass firings, ministry of 'labour', unions, bad marketing, etc.). Headcount will organically be reduced (retirements, getting a new job, etc.) combined with minimizing new hires, and boom! savings!!

theshackleford2• 1 year ago

They see that (e.g. with Notion-like-tools) if you enable the "AI", that thing can save _a lot_ of time on "Confluence-like" tasks.

If only it worked like this in reality. I used actual notion features literally this week and watched it fail so hard it was hilarious. It continually told me there was no documentation on X despite an entire page worth of documentation existing on it, had to be told this; at which point it apologised and regurgitated it.

Wow! What a time saver! I feel more productive already!

throwaway48476• 1 year ago

The white collar bloodbath is the jobs that could have been automated pre AI but weren't due to organizational inertia, corporate freedom building and an unwillingness to invest.

dottjt• 1 year ago

I think a huge tradeoff that people haven't mentioned is that in using AI to replace workers, you're introducing a dependency on AI that you previously didn't have. This poses a terrible long-term risk for companies.

bayareapsycho• 1 year ago

My last company (F50, ass engineering culture, pretends to be a tech company) went and fired all of the juniors at a certain level because "AI"

The funny part is, most of those juniors were hired in 2022-2024, and they were better hires because of the harsher market. There were a bunch of "senior engineers" who were borderline useless and joined some time between 2018-2021

I just think it's kind of funny to fire the useful people and keep the more expensive ones around who try to do more "managerial" work and have more family obligations. Smart companies do the opposite

indigoabstract• 1 year ago

So if I understand correctly, it's basically down between:

1. cure cancer

2. fix the economy

3. keep everybody happily employed.

And he's saying we can only pick two, or pick one. Except for the last one, that's not really an option.

boshalfoshal• 1 year ago

And as it stands, AI is nowhere close to (1) and (2), but is pretty close to making all of (3) redundant.

This could be because most work is actually frivilous (very possible), but its also easy for them to sell those since ostensibly (1) and (2) actually require a lot of out of distribution reasoning, thinking, and real agentic research (which current models probably aren't capable of).

(3) just makes the most money now with the current technology. Curing cancer with LLMs, though altruistic, is more unrealistic and has no clear path to immediate profitability because of that.

These "AGI" companies aren't doing this out of the goodness of their hearts with humanity in mind, its pretty clearly meant to be a "final company standing" type race where everyone at the {winning AI Company} is super rich and powerful in whatever new world paradigm shows up afterwards.

Warh00l• 1 year ago

pierceday.metalabel.com/aphone

The ‘white-collar bloodbath’ is all part of the AI hype machine

Related Stories