In the age of AI, don't let your skills atrophy

https://www.cyberdemon.org/2023/03/29/age-of-ai-skill-atrophy.html

by alnis • 3 years ago

86 97 3 years ago

nico4• 3 years ago

I propose the complete opposite.

Just surrender and enjoy the freedom.

Of course you can always choose, just like you can choose to remember people’s phone numbers and driving directions. But you don’t need it.

Syonyk2• 3 years ago

Freedom? You've got a weird definition of "freedom" if you think surrendering your autonomy to whoever happens to be running the services you rely on is "freedom."

The problem is that long term, you won't have any idea if what it's spitting out is reasonable or not. We already have more than enough problems with systems being too complex to understand. Solving that with "more complexity we can't understand" doesn't seem a great solution, personally.

13years4• 3 years ago

We already have more than enough problems with systems being too complex to understand.

Indeed, aligns with my own thoughts on the impossibility of AI alignment.

"However, each new layer and new tech that we add becomes a greater separation between what we can perceive and the complexity of the machine that becomes ever more incomprehensible. So we build ever more tech to help us understand the existing tech with each part being another component of complexity and potential problems."

https://dakara.substack.com/p/ai-singularity-the-hubris-trap

nico4• 3 years ago

That is already case.

Are you able to verify every single news article you read? Every source? Every journalist? All the info you read or hear?

You can’t, so you just trust. And let go of the complexity, you don’t need to consciously manage it and you are still ok.

Just like you trust your body to breath or to digest.

nonethewiser6• 3 years ago

Of course you can always choose, just like you can choose to remember people’s phone numbers and driving directions. But you don’t need it.

You do need to if your phone can’t store certain phone numbers. And you will be worse at it if you don’t practice.

Why can’t it store certain numbers? It’s impossible to know. Which phone numbers won’t it be able to store? It’s impossible to know.

msoad• 3 years ago

Much like the evolving layers of abstraction in technology, the necessity for individuals to master the intricate, underlying layers wanes as time progresses. The article brings to mind the analogy of Assembly programming; while there are undoubtedly experts in this domain whose contributions are indispensable, the majority of programmers can comfortably rely on higher-level abstractions without delving into the complexities of Assembly. As AI continues to advance, general-purpose programming appears to be following a similar trajectory. Individuals have the choice to either become one of the few specialists at that level or to embrace and harness the emerging abstractions, thereby unlocking greater efficiency and innovation in their endeavors. It is crucial to strike a balance, recognizing the importance of both mastery and leveraging AI to optimize our creative potential.

JohnFen• 3 years ago

while there are undoubtedly experts in this domain whose contributions are indispensable, the majority of programmers can comfortably rely on higher-level abstractions without delving into the complexities of Assembly.

True. You don't need to be an assembly expert to be a great programmer, but I do think you need to have a solid understanding of how computers work all the way down to the CPU level in order to be a great programmer. And once you have that knowledge, assembly isn't a big hurdle anyway.

TOMDM• 3 years ago

I don't think any individual needs to be an assembly expert, but as a community developers still need some people to be assembly experts.

Our systems will rot without people to maintain them, the AI aren't infallable yet.

The greater worry isn't that the general user will let mostly defunct skills rot, it's that we need a path to raise the next generation of experts in these niche skills.

fauxpause_3• 3 years ago

I don’t know anything about assembly and I can’t say I’ve ever thought about it.

livelielife• 3 years ago

just makes me want to combine compilers with understanding of LLMs and how chatGPT does what it does to make real next-generation (but really next level) programing languages.

LLMs will make distinctions like functional, dynamic, procedural, ... become obsolete just like all assembly languages are now just assembly, but in fact used to vary by architecture (and even by each architecture's generation/version)

assembly -> C -> something made with a compiler which embeds chatGTP

xg15• 3 years ago

Yeah, and you can thank those abstractions that today, we have Microsoft bragging that they got the Teams chat client down to only 9 seconds loading time on a machine with several multi-GHz cores and gigabytes of RAM. But that's all fine I guess, because it's good enough.

Good enough for what? Good enough that people will grudgingly use it because instantly quitting their job over it would be overblown. Their life will only become a little bit worse through it.

I'm frankly scared how AI-created "good enough" software will look.

nonethewiser6• 3 years ago

Teams is probably written in C# right? Are we really sure C# isn’t performant enough for desktop apps? Seems like a rather hot take.

Otherwise what is the non performant abstraction? I think the reality is just that teams is poorly done.

nonethewiser6• 3 years ago

I think about it in the same terms. We make better and better abstractions which makes it easier and easier to code. Usually with some performance loss that better hardware can cover.

The implication of that is this will have a similar impact as previous abstractions.

But two questions:

1. Does this scale?

2. Is the type of difference between Python and Assembly the same as between Python and AI Codegen?

muhammadusman• 3 years ago

from the post: What I’ve realized is that there are tasks I’m willing to let AI handle, as this enables me to take on more ambitious projects.

I've started doing this, getting some boilerplate or even throwaway code from ChatGPT helps me validate ideas quicker than if I wasn't using it.

13years4• 3 years ago

I am an existentialist: I am perfectly comfortable in a world with no inherent meaning. Sheer existence and action brings me joy. I spend hours of my day tinkering with computers not because of utility, but for the pleasure it brings.

I think if we could elicit this persons thoughts thoroughly we would find "meaning" underlying their perceived joy.

dmazin• 3 years ago

You can elicit my thoughts to your heart’s content. Of course there’s meaning. That’s different from inherent meaning (imbued into our world via a god or something).

13years4• 3 years ago

But that is not what the author stated. They didn't allude to a god like meaning. They cited utility as an example of meaning for which they had no association.

hgsgm3• 3 years ago

Generic expression, if not god.

Satisfying my animalistic/primatistic/humanist emotional urges is the meaning of my life.

ilaksh4• 3 years ago

I feel the same way I think. These days, I am happy when I am able to use GPT4 to complete a task faster or easier, but am also happy when I write some code "by myself".

However, as far as the conclusion of the article. This "age" started only a few months ago. If you are going to call it an age, then the conclusion doesn't hold up. In the long term, programming "by hand" is likely to be similar to wood carving today. It will be an artistic rather than utilitarian pursuit.

There are already multiple services and tools being built with the purpose of writing, deploying, and maintaining software in a 100% automated way using GPT3.5 and GPT4 and natural language specifications. I am building one of them.

We can't assume that these recent efforts to freeze the progress of AI will be successful. So we should anticipate very significant improvements in performance over the next few years.

Very shortly, everyone will realize that it is quite a huge waste of time and money to wait for a person to write code.

SoftTalker2• 3 years ago

Funny to hear that. I'm closer to 60 than 50, and it never occurs to me to use GPT to write my code. So I wonder if it's partly based on how long someone has been doing something. In retrospect, I never copied code from StackOverflow either (have used it occasionally to get ideas, but mostly felt the code I'd find there just wasn't very good, or at least was too different from my personal coding style to be something I'd want to use verbatim.

In the end I agree with you though -- if GPT code is "good enough" then paying people to write code will soon be looked at like using horse-drawn wagons to get from place to place.

aiappreciator2• 3 years ago

Old people do not like radical new technology. That's not a new observation.

It doesn't benefit you to adapt to the new workflow, with all its initial inefficiencies and errors, since you don't get the long term benefit.

ilaksh4• 3 years ago

I'm 45. Programming to some degree for 37 years. Same thing I always ask -- are you using GPT4 or 3.5? Can be a very big difference.

I think it really depends on what kind of programming you are doing as much as something like age. If you are trying to solve new problems in a new programming language or framework, then it will make more sense to get machine help than if you have a lot of experience in similar problems with the same language and libraries.

But that equation changes with GPT4 and especially beyond. Eventually you will probably try GPT4/5/ whatever and start using it.

digging• 3 years ago

I'm half your age and don't really find ChatGPT useful for my programming work. Maybe I'm not creative enough, but it often seems like more work to get ChatGPT to understand my problem than it is to solve the problem.

hgsgm3• 3 years ago

You don't even need antique examples like "wood carving".

Modern web app developers rarely write HTML by hand. They generate it from higher level language or templates.

ilaksh4• 3 years ago

Well.. lol.. guess I am not a modern web developer then because until GPT came out I wrote all of my HTML by hand, at least for the last several years. Before that I had phases with ASP/ASP.NET , WPF, and PHP.

rimliu• 3 years ago

And it shows :(

abledon• 3 years ago

Is every modern piece of software going to have `OPENAI_API_KEY` environment variable somewhere inside it ? uhoh

popotamonga• 3 years ago

Most of my customers already asking for it. They pay me i do it. Talk about job replacement, i'm raising rates by the day

hgsgm3• 3 years ago

It's the same when you use FOSS backup software that asks you to input your S3 API key or your remote host ssh key.

The only crucial difference is that we currently lack competition and self-ownes versions.

moonchrome4• 3 years ago

All this talk about chatgpt replacing programmers and skill atrophy - meanwhile all I'm getting out of chatgpt is bullshit and hallucinations. Copilot is amazing at boilerplate, but that's about it - I don't even read any suggestions that don't fall into that category anymore.

Copilot is amazing because it lets me stay in the flow when I need to churn out stuff (when I already know what I want). I would pay over 100$/month for a faster/less jittery copilot.

ChatGPT is cheap at 20$/month but not even worth that price.

8769780957897892• 3 years ago

all I'm getting out of chatgpt is bullshit and hallucinations

Are you using GPT3 or 4?

Ambitious1325• 3 years ago

I'm using 4 and I still am constantly having to babysit it and challenge it to get a working result out of it. Don't get me wrong it is absolutely saving me time, but it is very much like being the teacher of a very fast typing junior dev.

throwbadubadu2• 3 years ago

4 and full agree and still confusedly wondering.. I know I will be shouted down, but what are all you "tech guys" doing in your daily biz, just glueing hyperdocumented boilerplate together and spitting out the same textbook examples, I don't get it.

13years4• 3 years ago

That has been somewhat my experience as well. I found ChatGPT sometimes to useful to provide a hint for something I don't know how to do.

However, I never was able to get it to write a successful function for anything that would have been useful. It got it wrong every time.

majormajor2• 3 years ago

I think there's two camps of think-piece authors emerging - the ones who've tried a bunch of different examples of things and got passable-to-great-looking results; and the ones going deeper into specific areas and hitting the wall in terms of expertise and specificity. Using the GPT 4 API, I'm definitely often hitting limitations, especially around depth of information, and having to "prompt engineer" my way around them. After using a dozen prompt variants to try to prod it in the direction I want without seeing it reflect those changes, a bit of the magic wears off.

I'm bearish on the idea of long-term prompt engineering being a big skillset since I imagine the "understanding the prompt" side of the tools will get better, but I don't see it necessarily getting around the need for specificity of input. It feels like writing a task ticket and giving it to a junior person - what you get back might not be what you need, and a lot of time the true difficulty is knowing exactly what you need up front. Reducing that cycle time is wonderful, but doesn't replace the hard earned skills of knowing what to make.

fvdessen• 3 years ago

I am in the camp of people who see the current limitations but also see the rate of progress and think those limits may not stand for long. It is yet unclear if it will progress like autonomous driving or like playing go …

CuriouslyC• 3 years ago

Prompt engineer won't go away, it'll get more "engineer"-like. Knowing how to describe a point in a model's latent space for the generation you want is here to stay, but the black magic and art aspect of it will go away.

For example, in stable diffusion land, lots of people have intuition about the relationship between certain prompts and the output they produce. That intuition is embedding and training data specific, so it's not really transferrable (even to different fine tuned models for stable diffusion 1.5). However, I use clip interrogation to map the portions of the latent that my prompt is pointing to, evaluate the embedding text to find desirable/undesirable elements, then adjust the prompt or add negative prompts to navigate my generations towards what I want.

moonchrome4• 3 years ago

I agree with this, for me scenarios where I know what I want are better handled by copilot - I'm way better at writing code than gpt prompts - copilot then picks up the boilerplate as I go along - and since I know what I want it's easy to fact check.

There are some scenarios where it would be useful to have chat like interface in editor to prototype fast - hopefully copilot x delivers.

simonw3• 3 years ago

I think there are two camps of authors using. The ones who tried a bunch of different examples and got terrible results and wrote the whole thing off, and the ones who pushed on through that, kept exploring the capabilities of the model and couldn't believe how useful it could be once they figured out how best to use it.

jason-phillips• 3 years ago

meanwhile all I'm getting out of chatgpt is bullshit and hallucination

GiGo, basically

moonchrome4• 3 years ago

Doubt it, I've tried prompts where a simple Google query would lead to an answer to see if it could save me time (eg. query about AWS service usage, some azure stuff). In every instance I got a plausible sounding solution to my problem that would directly contradict documentation, it invents capabilities/features and misleads.

I've tried using it for code review on a few functions and tasked it to improve provided code - every time it would write worse code eg. I had some logic that would filter to a new list and then append replacement - it's refactor did filter -> add or replace for already filtered items, the reasoning was bullshit : fake performance claims about avoiding allocation when the "allocation" in question was value type, and the suggested alternative was replacing a vector with a hash map which is both logically wrong because of losing order, and slower for the use case.

For generating small stuff like a regex the pain you have to go through to get a correct prompt is higher than writing the thing and you still need to double check it.

I see no use case where chatgpt would improve my workflow in current stage and I've seen so many idiotic bugs recently when pressing the devs that introduced them it's basically "ChatGPT".

The one time it was useful was when I had to convert a model definition to open API spec - was easy to fact check and give feedback to get a decent solution.

chupaolo2• 3 years ago

I'm surprised with this response. I myself have found it extremely useful and ChatGPT has saved me tons of time with programming and non-programming tasks.

barbazoo• 3 years ago

all I'm getting out of chatgpt is bullshit and hallucinations
ChatGPT is cheap at 20$/month but not even worth that price.

This is so general obviously it's not true. It's providing lots of value to lots of people. To me this sounds like someone with the goal of confirming their own biases.

AverageDude2• 3 years ago

Yup. I always wonder why is my experience not like others? Are those PR people for Microsoft?

Example:

I gave chatgpt a list. Which looked like

st street

av avenue

Convert this to yaml format as

  st:
    name:
      street

And so on.

It failed spectacularly. Not even once but about 10 times. Even if it succeeded, it kept changing the output by doing ops which I never mentioned in the prompt (like reordering and merging duplicated values to a single key)

simonw3• 3 years ago

I have human intelligence and I can't figure out what the output you're hoping for here is.

Update: I tried that with GPT4 and got this:

    st:
      name: street
    av:
      name: avenue

GPT 3.5 didn't know what to do with it.

professoretc• 3 years ago

Even if it succeeded, it kept changing the output by doing ops which I never mentioned in the prompt (like reordering and merging duplicated values to a single key)

That's something I don't see mentioned enough; if you change the input to a LLM, that may potentially change the probabilities of all the output tokens. Most of us would be surprised if we told a junior developer to fix a bug in a specific module, and they submitted a PR which modified literally every file in the source tree, but that's entirely plausible with a LLM. Asking it to "fix" one thing may change/break completely unrelated things.

waboremo3• 3 years ago

In a way, people who are concerned about skills atrophying are repeating the cycle of previous generations, shaking their fist at higher level programming languages or drawing digital images. How dare you not keep the skills of those who came before you! Time will not be so kind to this view.

Some skills will atrophy, as article mentions we don't really care about road navigation anymore. But this is not a bad thing, we can focus our concerns elsewhere. Do you care exactly how your home is heated? Do you know how to maintain every part of your car? Do you spend every weekend keeping your skill on sewing up to date? Some of you might do all of these, but the point is we have let so many skills become things relegated to experts/tech that once used to be common.

It's this relegation that has allowed experts to become even deeper experts in their niche. So I don't worry about skills atrophying at all. There will always be a new crop of nerds (affectionately) to obsess over whatever niche can exist.

chupaolo2• 3 years ago

Exactly

mrbombastic• 3 years ago

Its not a bad thing until its a bad thing. There are certainly computer related tasks I am happy to relegate to AI, my concern is that we are already bad at maintaining software made by humans. If old school software engineering becomes an arcane art as everyone becomes prompt engineers and software starts to explode in complexity we better hope AI learns to maintain it too.

EarthLaunch• 3 years ago

Maybe it'll be like hard goods; a blender you buy today is made of moulded plastic and lasts 5 years, instead of being metal and made to last 20 years. For 1% of the price.

waboremo3• 3 years ago

Funny you mention old school software engineering, because it has become an arcane art! We do not feed punch cards into our computers anymore. Should also mention I don't believe "prompt engineer" will be a serious long term role, that task will just be incorporated into everyone's roles instead.

IMO maintaining software will become easier, so much of what we do today with maintenance is really just tedious. It's tedious to keep packages updated, to write documentation every update, to keep the community informed on what's going to happen in the next. It's tedious to rewrite the exact same functionality but in a modern language. List goes on, so much tedious busywork. What if these things were handled for core maintainers, so they can allocate funds to more complex problems?

nonethewiser6• 3 years ago

In a way, people who are concerned about skills atrophying are repeating the cycle of previous generations, shaking their fist at higher level programming languages or drawing digital images.

Maybe some.

But I resonate with the title because I noticed I started leaning on ChatGPT to avoid having to think through things. I found myself tweaking ChatGPT prompts and hoping for a better response when the previous one wasn’t good enough. Often a better response is not found, then I have to struggle to start thinking for myself.

Over time that leads to less repetition thinking through problems and more difficultly thinking through problems when you can’t use ChatGPT. Is that a good trade off?

mattcantstop• 3 years ago

I’ve been using AI to teach me concepts, and it’s great at it. It can sometimes be wrong, so having familiarity with the topic is important in letting it teach you. But OpenAI knows programming languages really well. It’s been amazing at teaching me concepts. Then I go test it to verify its teachings. It is certainly making me more productive at learning.

penjelly• 3 years ago

chatgpt is hit or miss but ive had success using it either way. The hard part will be not relying too much on it. We already know that some devs wont read error messages and instead will just jump into debugging, i trust thisll only get moreso with the advent of chatgpt. However, this might actually save those devs, assuming chatgpt answers correctly.

once you can upload images and ask it to answer based on that, itll be even more wild.

ducktective5• 3 years ago

I asked GPT-4 for the Imagemagick command to make the white parts of an image, semi-transparent.

It generated a command that made the fuzzy white parts [+1 on fuzzy] fully transparent [bad].

I told it that the result is not semi-transparent.

It apologized and gave me another command that produced a blank image. In another case a grayish image.

I told it this is not what I wanted, and it just looped here saying I'm sorry and giving me one of these above solutions.

As a matter of fact, this looping back and forth between half-working and non-working solutions is something that I've experienced every time when the first result was not what I asked...

Aside from the possibility of "emerging intelligence", I don't think this is the way to the AGI.

p1necone• 3 years ago

Aside from the possibility of "emerging intelligence", I don't think this is the way to the AGI.

My intuition is that this is the only way to create AGI. I don't think anyone is ever going to carefully intentionally construct an AGI, it's almost certainly going to emerge from something conceptually fairly simple.

I don't think it's impossible that our own brains are also basically just a big statistical prediction model too. Maybe AGI just requires our models to be 10/100/1000x as good. Or our training data needs to be broader in a qualitative way rather than a quantitative way that we haven't quite worked out yet.

ducktective5• 3 years ago

Can you prove Fermat last theorem via giving a high parameter LLM a very large amount of mathematical knowledge?

texuf• 3 years ago

Imagine you’ve only been a programmer for 2 years. Imagine not knowing what imagemagick is. Now get an answer that gets you 99% of the way to where you want to be. Now look at the documentation to see why the parameters aren’t doing what you want them to do. You just saved hours of work.

ducktective5• 3 years ago

Yes, GPT and LLMs got to be the most significant leap in the field of AI from 2010s onwards...It kinda "solves" NLP. It is the best human-language-based UI that humanity has every created.

But the point is, this thing does not understand what it's doing...it's become a cliche but the people who coined the term "statistical parrot" really knew what's up.

Like, I think there could be some GOFAI technique that may solve "figuring out how to use a tool like imagemagick/ffmpeg" mathematically, formally and deterministically.

And again, I'm all basing my view on the fact that this "emerging intelligence" is a mirage.

simonw3• 3 years ago

What did you conclude from this?

Was your conclusion "GPT-4 isn't any good for imagemagick commands" or "GPT-4 isn't useful for anything"?

ducktective5• 3 years ago

GPT-4 isn't good for tasks it was not trained on.

ilaksh4• 3 years ago

Did you try pasting in the relevant reference sections for those ImageMagick commands/options?

ducktective5• 3 years ago

No, I didn't know exactly what commands were useful. And I'm not subscribed to the pro version so pasting the whole documentation would cost a lot.

killjoywashere• 3 years ago

I feel like, if anything, AI will push people to upskill. A) it's obviously necessary, B) you will increasingly find yourself increasingly identifying edge cases, C) if you're in an exposed field (e.g. diagnostic medicine) you will find fewer people coming into the field, so there won't be a lot of backfills in the workforce.

comment_ran• 3 years ago

With ChatGPT, it’s too easy to implement ideas without understanding the underlying concepts or even what the individual lines in a program do. Doing so for unfamiliar tasks means failing to learn something new, while doing so for known tasks leads to skill atrophy.

Imagine this: A hypothetical "GPT 7" tech can effortlessly create a starship capable of shuttling you from Earth to Mars. It's so reliable that after 100 uses, it never fails. Perfect flights, precise landings, and zero issues. All it takes is a simple command prompt "design a starship to Mars". In this scenario, is there a need to learn the intricacies of aerodynamics, gravitational forces, or rocket science?

The notion here is to liberate ourselves from limitations and focus on what truly interests us. That's the end game.

cheald• 3 years ago

Someone has to. The "humanity designed tech that it became reliant on and forgot how to reimplement, so when it broke down nobody could fix it" is a recurring scifi trope.

m0llusk• 3 years ago

Frank Herbert's Destination: Void arguably takes one through such a scenario, a highly recommended read!

dahwolf• 3 years ago

"The thing I’m most excited about in our weird new AI-enhanced reality is the way it allows me to be more ambitious with my projects. In the past I’ve had plenty of ideas for projects which I’ve ruled out because they would take a day—or days—of work to get to a point where they’re useful. I have enough other stuff to build already!"

If these projects are somehow meaningful to you on a personal level or in a small circle, great. However, let's not look past the fact that pretty much every human will achieve this super productivity. Hence, in a broader sense most projects will not be interesting, competitive...or more likely...looked at because of sheer abundance.

The other thought I'm having is regarding the rate of change in mastering AI. It's frankly inhumane. You can work in tech and thrive on a tech stack for some 5 years, sometimes 10. Working the AI...whatever you learn is outdated the next week.

In the age of AI, don't let your skills atrophy

Related Stories