gensym

I fear they are laying the groundwork to ban US citizens from using Chinese models, so that they can make sure that Brockman gets his money's worth.

I get what you're saying and I see that how my comment could be interpreted. I wish I had worded it better.

The most vibrant places in the world (again, IME) have a diversity of abilities, backgrounds, ages, and economic status.

There's another type of place - the sort of place I had in mind - that doesn't. That attracts people who are trying to get away from law and any sort of social contract. Some of those places look attractive on the surface. That's the sort of place I'm talking about. They're probably not as common as they seem to be to me.

Also, IME, places with no employers tend to be populated primarily by addicts, disability fraudsters, and other criminals.

Gotta fuckin love hn, where you can't say the simplest thing without getting thrown into a well, actually.

I wasn't claiming to have run a damned study. Just making an observation. The thing is, people who don't enjoy exercising but start exercising anyway tend to find they enjoy it after a while. Saying people enjoy exercise is only slightly more controversial than saying "ice cream tastes good". There's a bit of activation energy required, but it's not a huge stretch to notice that something that is good for our bodies gives us pleasure.

What a weird question. The amount depends on what you hope to accomplish.

Want to be able to get on the floor and play with your grandkids? That's going to require a certain amount of mobility work.

Want to be able to hike up Mount Whitney? You'll need regular cardio.

I also hate the phrasing that exercise is something to "get away with". Most people who exercise regularly enjoy it. Moving your body is fun, especially once you get fit enough to do it without pain or immediate exhaustion.

Favored companies get access to frontier models, which gives them a competitive advantage, starving out smaller companies. Any smaller companies that do manage to innovate ultimately get acquired by the favored companies since they are worth more with access to frontier models than without (which is effectively a discount on the purchase price of those companies).

I'm not saying that's going to happen, but it is one possible scenario that, over time, would be disastrous for innovation and freedom.

Just because a policy is legal doesn't mean it cannot be disastrous.

So you think Anthropic is using internal AI assistants to pull away from competitors and the leapfrogging we've seen over the last several years is now done?

That seems to me to be the most concrete and least obvious prediction in the quoted text.

I don't think that's happening. If that were generally accepted as true I would expect OpenAI to be unable to successfully IPO.

I am not cynical enough to believe that Anthropic's warnings are pure marketing hype.

Nor am I. I think they believe that AI poses a grave danger, and they are playing the prisoner's dilemma as an unvirtuous actor.

1. If anyone builds strong AI, it may be catastrophically bad.

2. If anyone builds strong AI, it will be better for the builder than for anyone who does not. Either because it won't be catastrophically bad so the builder will get to enjoy all the spoils indefinitely or because it will and at least the builder will be rich for a while.

I grew up in Illinois in the 80s and 90s, so I had a front-row seat to that gutting. Watching my dad move from factory to factory, go to night school to learn tool-and-die and then see all the tool-and-die jobs move overseas.

So no, it didn't stop it. But the fact that it didn't happen overnight did allow my dad to feed his family and make sure my brother and I had better opportunities. If that gutting had happened over a couple years rather than a couple decades, we would have starved, and the knowledge that we did it "ourselves" rather than have it forced upon us by globalization wouldn't have stocked our pantry.

So, yes, there are direct parallels here. I want to give US families a chance to adapt.

I don't understand that this argument. Why does the net improvement of the technologies listed imply that AI will also have a net improvement? Are you just arguing that there's no such thing as technology that is harmful on net?

For people worried about their livelihoods, there's value in slowing AI adoption to give our economy time to transition rather than just throwing a lot of people out of work all at the same time.

Not a single social media restriction experiment has included people under the age of 16. We do not know how social media bans will affect the young people being targeted by them because we have never tested this with them!

I've also never tested my ability to survive a 100ft fall. Maybe I can! We have no way of knowing!

Virtually all schools in the United States report that they use social media for communications, including for key announcements such as making families aware of upcoming opportunities, educational programming, and key deadlines. The reliance on social media for communication and resource sharing, while banning youth from these same platforms, sends mixed messages to young people and limits their access to health promoting information and resources.

That's a good point. There's no other way that schools could communicate such things. My childhood in the 80s and 90s certainly didn't include Scouts, 4-H, Band, Drama, Cross-Country, etc! I'm sure with social media bans for youth, schools will just continue to use social media to try to communicate to kids rather than adapting.

I have to assume the authors of this paper know how dumb it is and just don't care since most people will only read the headline.

I can't speak for anyone else, but I find AI to be very effective. It can do nearly all coding tasks many OOMs faster than I can. And I'm able to get it to produce high-quality code in the process. Using AI, our codebase actually has less tech debt than any time I can remember. I would be less effective if I wasn't using AI and if I wasn't finding new ways to leverage it.

That doesn't mean I enjoy it using AI. I loved coding. I was really good at it! I spent decades honing my abilities, and while some of those skills are still applicable when working with AI, many are not.

I want my company to be successful, so I work as effectively as I can. Unfortunately, the most effective method of working no longer scratches the creative/craftsman itch that it used to.

Right now I'm mitigating it by taking up creative writing in the evenings. That's difficult and creative in a way that coding used to be. Identifying and solving character and story problems feels like debugging and designing used to. Learning to craft effective prose feels like it used to when I was picking up a new programming language and learning its idioms.

The map is not the territory

It’s pretty simple. People think that AI will take their jobs and maybe murder them, probably because the people developing AI have said it’s going to take their jobs and maybe murder them.

Opposing data centers is the biggest lever most people have to impede AI development.

So it's not where you buy those shirts that say "Female Body Inspector?"

Yeah, it makes me wonder about their planned rollout to more of Southern California, where flooded roads aren't uncommon, especially in some of the valleys.

I think this disconnect is based on the ambiguity in the term "AI".

"AI" as tech - the models, how to train them, etc. Isn't going to go anywhere short of a Library-of-Alexandria-type catastrophe. We know how to do it and it's useful, so why would we forget?

However, "AI" as the thing that is enveloping our culture - the slop everywhere, the mandates to use it at work regardless of its usefulness, the constant talk about it being the future, the machine-dominated future that's been promised/threatened by the heads of the labs - we do still have a chance to put that onto the scrapheap.

We read a bunch of content online and then use it in our work product.

We also have societal norms around plagiarism.

Additionally, the claim that because people have the right to do something then we should extend that right to machines is strong. (And one I certainly reject).

We started aggressively adopting AI in my company this year. I think I disliked (though never hated) it for a few days, but it’s a systemic change that I can’t just push back against.

I'm right there with you. I think AI will be bad as a whole for the world, but I use it for work every day and am pushing my team to use it more. I think it's a really effective tool for my company even if it's going to be bad for the world overall.

I don’t believe that strong public opinion can stop technological development either—just take nuclear for example.

I see nuclear as an example of where public opinion did stop development. In the US at least, we've basically given up on nuclear power, much to our detriment.

Another example of this is human cloning, which seemed inevitable back when Dolly the sheep was first cloned.

Were you around in the 90s? People really were more optimistic then, at least in the US. It wasn't perfect, but it really did feel like things were getting better. The Clinton administration had to start doing studies about whether paying off the national debt would be globally destabilizing! We really were talking about the "end of history". We thought the Internet would bring people together and end bigotry.

It's funny how easily you can differentiate people in the tech industry who spend all their time with others in the tech industry from those who don't.

The former either seem puzzled about the general public's anger at AI or dismissive of it ("they don't really hate it - look at ChatGPT usage!", "they only hate it because they've been misled about water usage!" and so on).

Non-techies aren't as stupid as people in the tech industry think. Normies can see their social media feeds filling up with slop. They see people in their social circles who can no longer hold a normal conversation without feeding everything into ChatGPT. And - most importantly, I suspect - they are seeing the plan they've built their lives around - get your kids to do well in school, get them into college so they can have a good career and make enough to pay of the loans that plan will require - being casually dismissed by AI boosters ("they'll be plenty of jobs, we just don't know what they are yet!").

Here's a clue for people who don't understand the backlash: if you don't understand that stability has value on its own, then you lack a basic understanding of what more people actually care about.

Yeah, I guess it was just charity that led them to develop a really fast, efficient processor and to put good memory in their machines in the first place.

Don't mistake not caring about "specs" with being indifferent to the experience.

I've been thinking about this and have wanted to discuss it with people. I think the 10x thing has been broken, but I don't think it's because the premise of "No Silver Bullet" was false - I think it's because LLMs have the ability to navigate some of the _essential_ complexity of problems.

I don't think anyone has really wrestled with the implications of that yet - we've started talking about "deskilling" and "congnitive debt" but mostly in the context of "programmers are going to forget how to structure code - how to use the syntax of their languages, etc et etc)." I'm not worried about that as it's the same sort of thing we've seen for decades - compilers, higher-order languages, better abstracts, etc etc etc.

The fact that LLMs are able to wrestle with essential complexity means that using them is going to push us further and further from the actual problems we're trying to solve. Right now, it's the wrestling with problems that helps us understand what those problems are. As our organizations adopt LLMs that are able to take on _those_ problems - that is, customer problems, not problems of data, scaling, and so forth - will we hit a brick wall where we lose that understanding? Where we keep shipping stuff but it gets further and further from what our customers need? How do we avoid that?

Any sort of "X Literacy" raises red flags for me. Actual _literacy_ - as in, being able to critically read and comprehend stuff of sufficient complexity - is basically a superpower that makes learning all these other skills possible, and it seems to be in terribly short supply.

What a strange sentiment.

The world would be better without nuclear technology! Yes, it can be used to generate electricity in a much cleaner way than other technologies, but its potential to kill so many people has shaped geopolitics for generations. If we could get the energy without the destruction, that would be ideal, of course. But there's an unacceptable chance that nuclear weapons will kill everyone I care about.

I think AI is actually similar. It's really useful for all sorts of stuff - I'm running multiple agents writing code as we speak. But it's also going to make the world much more dangerous by giving people the ability to kill and spy like never before.

There are better arguments than the water thing.

Honestly, it's weird to me how fixated both sides are on water.

People against data centers overestimate water usage, but people who think we should build as many as we can, as fast as we can seem the think that "actually they don't use that much water" somehow negates the more real issues with them.

Unfortunately, right now, it seems like history belongs to the people who bullshit.

Yeah, I agree with you. There is a right way to do that, and I assumed the comment I was replying to was describing the wrong way to do it given their reaction.

I've ended plenty of interviews early when it's clear the candidate isn't going to work out. I agree there's no point in wasting everyone's time, and hiring is time consuming enough. But there's a way to do it with kindness, and I think everyone in the interviewer's chair should have some sense of how. (That said, there are some candidates that are going to take rejection poorly no matter what - you can control how you treat a candidate but now how they react).