[HN Gopher] Avoiding skill atrophy in the age of AI
___________________________________________________________________
Avoiding skill atrophy in the age of AI
Author : NotInOurNames
Score : 320 points
Date : 2025-04-25 08:30 UTC (1 days ago)
(HTM) web link (addyo.substack.com)
(TXT) w3m dump (addyo.substack.com)
| rekado wrote:
| I really dislike the cartoons, because they are carelessly
| generated images. On the first look they appear to be actual
| cartoons (you know, where details were deliberately placed to
| convey meaning), but the more you look the more confusing they
| get because it seems that most details here are accidental.
|
| To me bad illustrations are worse than no illustrations. They
| also reflect poorly on the author, so I'm much less inclined to
| give them the benefit of the doubt, and probably end up
| dismissing their prose.
| jeremyleach wrote:
| Which only goes to emphasise the point the author makes. Over-
| reliance on AI, in this case, for image generation.
| mtsolitary wrote:
| Seems like AI was leaned on for the text as well...
| nottorp wrote:
| But maybe the author manually reviewed every word :)
| lemonberry wrote:
| Given that he's a published author and has been writing
| publicly for years, I'd love to hear if and how he uses AI
| for his writing.
| mathgeek wrote:
| There is a certain sense of "the leopards won't eat my face"
| that crosses my mind every time someone writes about skills in
| the age of AI but then inserts generated images.
| hk__2 wrote:
| For anyone like me who didn't know what this "the leopards
| won't eat my face" refers to: https://en.wikipedia.org/wiki/T
| urkeys_voting_for_Christmas#:...
| true_religion wrote:
| Where there is AI illustrations today, in the past would be
| clip art with little relevance to the work.
| gherkinnn wrote:
| I've been using Claude to great effect to work my way through
| ideas and poke holes in my reasoning. Prompting it with "what am
| I missing?", "what should I look out for?" and "what are my
| options?" frequently exposes something that I did miss. I need to
| be the architect and know what to ask and know what I don't know.
| Given that, Claude is a trusty rubber duck at worst and a
| detective at best.
|
| It then suggests a repository pattern despite the code using
| active directory. There is no shortcut for understanding.
| bgwalter wrote:
| The average IQ will probably drop at least ten points in the next
| ten years, but everyone will write (AI generated) blog posts on
| how their productivity goes up.
| lordofgibbons wrote:
| People have been afraid of the public getting dumber since the
| start of mass book printing and has happened with every
| following new technology since then.
| hk__2 wrote:
| "In the age of endless books, we risk outsourcing our
| thinking. Instead of grappling with ideas ourselves, we just
| regurgitate what we read. Books should be fuel, not crutches
| --read less, think more."
|
| Or even: "In the age of cave paintings, we risk outsourcing
| our memory. Instead of remembering or telling stories, we
| just slap them on walls. Art should be expression, not escape
| --paint less, live more."
| bgwalter wrote:
| Cave paintings were made by AI robots trained on the IP of
| real painters?
| hk__2 wrote:
| How they are made is irrelevant to the point.
| fvdessen wrote:
| Unfortunately research shows that nowadays we're actually
| getting dumber, literacy rates are plummeting in developed
| countries.
|
| [1] https://www.oecd.org/en/about/news/press-
| releases/2024/12/ad...
| blackoil wrote:
| Do you mean developed? OECD are all rich western countries.
| fvdessen wrote:
| Yes, sorry, corrected
| looofooo0 wrote:
| Is this culture based or reproduction based?
| bgwalter wrote:
| The IQ in the US started declining since the start of the
| Internet:
|
| https://www.popularmechanics.com/science/a43469569/american-.
| ..
|
| "Leading up to the 1990s, IQ scores were consistently going
| up, but in recent years, that trend seems to have flipped.
| The reasons for both the increase and the decline are sill
| [sic!] very much up for debate."
|
| The Internet is relatively benign compared to cribbing
| directly from an AI. At least you still read articles, RFCs,
| search for books etc.
| jvanderbot wrote:
| As someone who grew up reading encyclopedias, LLMs are the
| most interesting invention ever. If Wikipedia had released
| the first chat AI we'd be heralding a new age of knowledge
| and democratic access and human achievement.
|
| It just so happens unimaginative programmers built the
| first iteration so they decided to automate their own jobs.
| And here we are, programmers, worrying about the dangers of
| it all not one bit aware of the irony.
| cess11 wrote:
| As someone who grew up reading encyclopedias, I find
| LLM:s profoundly hard to find a use for besides crude
| translations of mainly formal documents and severely
| unreliable transcriptions.
|
| I like structured information and LLM:s output
| deliberately unstructured data that I then have to vet
| and sift out and structure information from. Sure, they
| can fake structure to some extent, I sometimes get XML or
| JSON that I want, but it's not really either of those and
| also common that they inject subtle, runny shit into the
| output that take longer to clean out than it would have
| to write a scraper against some structured data source.
|
| I get that some people don't like reading documentation
| or talking to other people as much as having a fake
| conversation, or that their editors now suggest longer
| additions to their code, but for me it's like hanging out
| with my kids except the LLM is absolutely inhuman,
| disgustingly subservient and doesn't learn. I much prefer
| having interns and other juniors around that will also
| take time to correct but actually learn and grow from it.
|
| As search engines I dislike them. When I ask for a subset
| of some data I want to be sure that the result is
| exhaustive without having to beg for it or make threats.
| Index and pattern matching can be understood, and come
| with guarantees that I don't just get some average or
| fleeting subset of a subset. If it's structured I can
| easily add another interactive filter that renders
| immediately. They're also too slow for the kind of non-
| exhaustive text search you might use e.g. Manticore or
| some vector database for, things like product
| recommendations where you only want fifteen results and
| it's fine if they're a little wonky.
| ladeez wrote:
| Yeah doesn't matter what you prefer. New hardware will
| boot strap models and eliminate the layers of syntax
| sugar devs use to write and ship software.
|
| Hardware makers aren't living some honorific quest to
| provide for SWEs. They see a path to claim more of the
| tech economy by eliminating as many SWE jobs as possible.
| They're gonna try to capitalize on it.
| lazide wrote:
| Bwahaha. This is about as likely (in practice) as the
| whole 'speech to text software means you'll never need to
| type again' fad.
| pyrale wrote:
| Before you jump to conclusions, you should make a
| reasonable claim that IQ is still a reasonable measure for
| an individual's intellectual abilities in this context.
|
| One could very much say that people's IQ is bound to
| decline if schooling decided to prioritize other skills.
|
| You would also have to look into the impact of factors
| unrelated to the internet, like the evolution of schooling
| and its funding.
| rahimnathwani wrote:
| IQ scores may be declining, but it's far from certain that
| the thing they're trying to measure (g, or general
| intelligence) have actually declined.
|
| https://open.substack.com/pub/cremieux/p/the-demise-of-
| the-f...
| tptacek wrote:
| That's an article apparently from a white nationalist,
| Jordan Lasker, a collaborator of Emil Kirkegaard's. For a
| fun, mathematical take (by Cosma Shalizi) on what
| statistics tells us about "g" itself:
|
| http://bactra.org/weblog/523.html
| rahimnathwani wrote:
| That's an article apparently from a white nationalist,
| Jordan Lasker, a collaborator of Emil Kirkegaard's.
|
| Do you have any comments about the article itself?
| http://bactra.org/weblog/523.html
|
| Thanks! I read the introduction, and will add it to my
| weekend reading list.
|
| The author objects to treating 'g' as a causal variable,
| because it doesn't help us understand how the mind works.
| He doesn't deny that 'g' is useful as a predictive
| variable.
| tptacek wrote:
| I highly recommend reading the whole piece.
| rahimnathwani wrote:
| I will! Weekend starts soon!
| tptacek wrote:
| The Borsboom and Glymour papers he links to are worth a
| skim too. It's a really dense (in a good way!) piece.
| Also shook up the way I think about other psych findings
| (the "big 5" in particular).
| Gigachad wrote:
| Pretty good chance that this is the impact of a generation
| of lead poisoned children growing up with stunted brains.
| qntmfred wrote:
| Plato wrote in Phaedrus
|
| This invention [writing] will produce forgetfulness in the
| minds of those who learn to use it, because they will not
| practice their memory. Their trust in writing, produced by
| external characters which are no part of themselves, will
| discourage the use of their own memory within them.
| nottorp wrote:
| That's okay, we're moving to post reading :)
| dockercompost wrote:
| What did you say?
| nottorp wrote:
| I'll have an AI make a tiktok video to summarize my post
| just for you!
| dockercompost wrote:
| Thanks! I'll ask NotebookLM to make a podcast out of it!
| Aeolun wrote:
| I'm inclined to believe he was right? There's other
| benefits to writing (and the act of writing) that weren't
| well understood at the time though.
| whatnow37373 wrote:
| He was not wrong. We forget stuff all the time and in huge
| quantities. I can't even remember my own phone number half
| of the time.
|
| Those guys could recite substantial portions of the Homeric
| epics. It's just that there is more to intelligence than
| rote memorization. That's the good news.
|
| The bad news is that this amorphous "more" was "critical
| thinking" and we are starting to outsource it.
| namaria wrote:
| Writing had existed for 3000 years by then, alphabetic
| writing in Greek had existed for several centuries. The
| quote about "the invention of writing" is Socrates telling
| a story where a mythical Egyptian king says that.
|
| Socrates also says in this dialogue:
|
| "Any one may see that there is no disgrace in the mere fact
| of writing."
|
| The essence of his admonishment is that having access to
| written text is not enough to produce understanding, and I
| not only tend to agree, I think it is more relevant than
| ever now.
| Aeolun wrote:
| We've probably compensated by ease of information
| dissemination. We've pretty much reached the peak of that
| now, so the only thing we can do is dumb shit down further?
|
| Maybe someone can write one of those AI apocalypse novels in
| which the AI doesn't go off the rails at all but is instead
| integrated into the humans such that they become living
| drones anyhow.
| hk__2 wrote:
| I feel like the whole blog post could have been written 10-20
| years ago if you replace "AI" with "Google".
| XorNot wrote:
| Could've been written any time in the last 2500 years
| really...and has been[1]
|
| [1] https://slate.com/technology/2010/02/a-history-of-media-
| tech...
| godelski wrote:
| Yes. But did you read the article? The advice would still be
| good with Google and there's certainly a lot of programs who
| see their jobs as gluing together stack overflow code. It's
| even true that you should struggle a little before reaching for
| the manual! (It's just slower so you'll probably think a little
| while finding the right page)
|
| The blog really says the same thing that's told in any
| educational setting: struggle a like first. Work your brain.
| Don't instantly reach for help when you don't know, try first,
| then reach out.
|
| The difference with the llm is the scale and ease of being able
| to reach out. Making people use it too early and too often.
| hk__2 wrote:
| Agreed. (of course I read the article, otherwise I couldn't
| have got this feeling)
| heymax054 wrote:
| I just use Anki (spaced repetition) to learn new concepts. Now
| with AI, one added benefit is "avoiding skill/knowledge atrophy"
| the more I use LLMs to generate the clde.
| dmazin wrote:
| Yeah, I picked up Anki and ChatGPT the same year. I _heavily_
| use both. I 'd say that Anki has increased my intelligence far
| more than LLMs have. Anki means that I no longer forget
| interesting things I learn day to day. To me, that's more
| valuable than the fact that LLMs make me code a multiple
| faster.
| gchamonlive wrote:
| We can finally just take a photo of a textbook problem that has
| no answer reference and no discussion about it and prompt an LLM
| to _help us understand_ what 's missing in our understanding of
| the problem, if our solution is plausible and how we could verify
| it.
|
| LLM changed _nothing_ though. It 's just boosting people's
| _intention_. If your intention is to learn, you are in luck! It
| 's never been easier to teach yourself some skill for free. But
| if you just want to be a poser and fake it until you make it, you
| are gonna be brainrot waaaay faster than usual.
| nottorp wrote:
| > of a textbook problem
|
| Well said. Textbook problem that has the answer everywhere.
|
| The question is, would you create similar neural paths if
| reading the explanation as opposed to figuring it out on your
| own?
| hnthrowaway0315 wrote:
| I believe there is a lot of value to trying to figure out
| things by myself -- ofc only focusing on things that I really
| care for. I have no issue relying on AI on most of the work
| stuffs, they are boring anyway.
| codr7 wrote:
| I personally can't thing of anything more boring than
| verifying shitty, computer generated code.
| bsaul wrote:
| i'm using chatgpt for this exact case. It helps me verify my
| solution is correct, and when it's not, where is my mistake.
| Without it, i would have simply skipped to the next problem,
| hoping i didn't make a mistake. It's definitely a win.
| spiritplumber wrote:
| I mostly use chatgpt to make my writing more verbose
| because I've been told that it's too terse.
| lee-rhapsody wrote:
| Terse writing is a gift. I'm an editor and I wish my
| writers were more terse.
| MonkeyClub wrote:
| > would you create similar neural paths
|
| Excelent point, and I believe the answer is a resounding
| negative.
|
| Struggling with a problem generates skills and knowledge
| which you then possess and recall more easily, while reading
| an answer merely acquires some information that competes with
| a whole host of other low-effort information that you need to
| remember.
| netdevphoenix wrote:
| Unlikely. Reading the explanation involves memorising it
| temporarily and at best understanding what it means at a
| surface level. Figuring it out on your own also involves
| using and perhaps improving your problem solving skills in
| addition to understanding the explanation at a deeper level.
| I feel LLMs will be for our reasoning skills what writing was
| for our memory skills.
|
| Plato might have been wrong about the ills of cyberization
| cognitive skill such as memory. I wonder if two thousand
| years later from then, we will be right about the ills of
| cyberization of a cognitive skill such as reasoning
| namaria wrote:
| > Reading the explanation involves memorising it
| temporarily and at best understanding what it means at a
| surface level.
|
| I agree. I don't really feel like I know something unless I
| can go from being presented with a novel instance of a
| problem in that domain and work out a solution by myself,
| and also explain that to someone else - not just happen
| into a solution.
|
| > Plato might have been wrong about the ills of
| cyberization cognitive skill such as memory.
|
| How so? From the dialogue where he describes Socrates
| discussing writing I get a pretty nuanced view that lands
| pretty much where you did above: access to writing fosters
| a false sense of understanding when one can read
| explanations and repeat them but not actually internalize
| the reasoning behind it.
| gchamonlive wrote:
| What's the difference? Isn't explaining things so that people
| don't have to figure out by themselves the whole point of the
| educational system?
|
| You will still need the textbook because llms hallucinate
| just as much as a teacher can be wrong in class. There is no
| free lunch, llm is just a tool. _You create the meaning_.
| skydhash wrote:
| > _What 's the difference? Isn't explaining things so that
| people don't have to figure out by themselves the whole
| point of the educational system?_ THEN SAID
| A teacher, Speak to us of Teaching. And he
| said: No man can reveal to you aught but that
| which already lies half asleep in the dawning of your
| knowledge. The teacher who walks in the shadow
| of the temple, among his followers, gives not of his wisdom
| but rather of his faith and his lovingness. If
| he is indeed wise he does not bid you enter the house of
| his wisdom, but rather leads you to the threshold of your
| own mind. The astronomer may speak to you of his
| understanding of space, but he cannot give you his
| understanding. The musician may sing to you of
| the rhythm which is in all space, but he cannot give you
| the ear which arrests the rhythm nor the voice that echoes
| it. And he who is versed in the science of
| numbers can tell of the regions of weight and measure, but
| he cannot conduct you thither. For the vision of
| one man lends not its wings to another man. And
| even as each one of you stands alone in God's knowledge, so
| must each one of you be alone in his knowledge of God and
| in his understanding of the earth.
|
| The Prophet by _Kahlil Gibran_
| atoav wrote:
| My impression is similar. LLMs are a godsend for those willing
| to learn, as they can usually answer extremely specific
| questions well enough to at least send you into the right
| general direction.
|
| But if you're so insecure about yourself that you invest more
| energy into faking it than other people do in actually doing it
| this, this is probably a one-way street to never actually be
| able to do anything yourself.
| cube2222 wrote:
| > It's just boosting people's intention.
|
| This.
|
| It will in a sense just further boost inequality between people
| who want to do things, and folks who just want to coast without
| putting in the effort. The latter will be able to coast even
| more, and will learn even less. The former will be able to
| learn / do things much more effectively and productively.
|
| Since good LLMs with reasoning are here, I've learned so many
| things I otherwise wouldn't have bothered with - because I'm
| able to always get an explanation in exactly the format that I
| like, on exactly the level of complexity I need, etc. It brings
| me so much joy.
|
| Not just professional things either (though those too of
| course) - random "daily science trivia" like asking how exactly
| sugar preserves food, with both a high-level intuition and low-
| level molecular details. Sure, I could've learned that if I
| wanted too before, but this is something I just got interested
| in for a moment and had like 3 minutes of headspace to dedicate
| to, and in those 3 minutes I'm actually able to get an LLM to
| give me an excellent tailor-suited explanation. This also made
| me notice that I've been having such short moments of random
| curiosity constantly, and previously they mostly just went
| unanswered - now each of them can be satisfied.
| namaria wrote:
| > Since good LLMs with reasoning are here
|
| I disagree. I get egregious mistakes often from them.
|
| > because I'm able to always get an explanation
|
| Reading an explanation may feel like learning, but I doubt
| it. It is the effort of going from problem/doubt to
| constructing a solution - and the explanation is a mere
| description of the solution - that _is_ learning. Knowing
| words to that effect is not exactly learning. It is an
| emulation of learning, a simulacrum. And that would be bad
| enough _if_ we could trust LLMs to produce sound explanations
| every time.
|
| So not only getting the explanation is a surrogate of
| learning something, you also risk internalizing spurious
| explanations.
| smallnix wrote:
| I think so too. Otherwise every Google maps user would be
| an awesome wayfinder. The opposite is true.
| myaccountonhn wrote:
| Every now and then I give LLMs a try, because I think it's
| important to stay up to date with technology. Sometimes
| there have been specs that I find particularly hard to
| parse in domains I am a bit unfamiliar in where I thought
| the AI could help. At first the solutions seemed correct
| but then on further inspection, no they were far more
| convoluted than needed, even if they worked.
| FridgeSeal wrote:
| I can tell when my teammate's code contains LLM-
| induced/written code, because it "functionally works" but
| does so in a way that is so overcomplicated and unhinged
| that a human isn't likely to have gone out of their way
| to design something so wildly and specifically weird.
| skydhash wrote:
| That's why I don't bother with LLMs even for scripts.
| Scripts are short for a reason, you only have so much
| time to dedicate on it. And often you pillage from one
| script to use in another, because every line is doing
| something useful. But almost everything I generated with
| LLM are both long and full of abstractions.
| cube2222 wrote:
| First, as you get used to LLMs you learn how to get
| sensible explanations from them, and how to detect when
| they're bullshitting around, imo. It's just another skill
| you have to learn, by putting in the effort of extensively
| using LLMs.
|
| > Reading an explanation may feel like learning, but I
| doubt it. It is the effort of going from problem/doubt to
| constructing a solution - and the explanation is a mere
| description of the solution - that is learning. Knowing
| words to that effect is not exactly learning. It is an
| emulation of learning, a simulacrum. And that would be bad
| enough if we could trust LLMs to produce sound explanations
| every time.
|
| Every person learns differently, and different topics often
| require different approaches. Not everybody learns exactly
| like you do. What doesn't work for you may work for me, and
| vice versa.
|
| As an aside, I'm not gonna be doing molecular experiments
| with sugar preservation at home, esp. since as I said my
| time budget is 3 minutes. The alternative here was reading
| about it on wikipedia or some other website.
| namaria wrote:
| > It's just another skill you have to learn, by putting
| in the effort of extensively using LLMs.
|
| I'd rather just skip the hassle and keep using known good
| sources for 'learning about' things.
|
| It's fine to 'learn about' things, that is the extent of
| most of my knowledge. But from reading books, attending
| lectures, watching documentaries, science videos on
| youtube or, sure, even asking LLMs, you can at best
| 'learn about' things. And with various misconceptions at
| that. I am under no illusion that these sources can at
| best give me a very vague overview of subjects.
|
| When I want to 'learn something', actually acquire
| skills, I don't think that there is any other way than
| tackling problems, solving them, being able to build
| solutions independently and being able to explain these
| solutions to people with no shared context. I know very
| few things. But I am sure to keep in mind that the many
| things I 'know about' are just vague apprehensions with
| lots of misconceptions mixed in. And I prefer to keep to
| published books and peer reviewed articles when possible.
| Entertaining myself with 'non-fiction' books, videos etc
| is to me just entertainment. I never mistake that for
| learning.
| jerkstate wrote:
| Reading an explanation is the first part of learning,
| chatgpt almost always follows up with "do you want to try
| some example problems?"
| Phanteaume wrote:
| Some problems do not deserve your full attention/expertise.
|
| I am not a physicist and I will most likely never require
| to do anything related to quantum physics in my daily life.
| But it's fun to be able to have a quick mental model to
| "have an idea" about who was Max Planck.
| sethammons wrote:
| I used chatgpt to get comfortable with DIYing my pool filter
| work. I started clueless "there is a thing that looks like
| $X, what is it" to learning I own a sand filter and how to
| maintain it.
|
| My biggest barrier to EVERYTHING is not knowing the right
| word or term to search. LLMs ftw.
|
| A proper LLM would let me search all of my work's artifacts
| when I ask about some loose detail I half remember. As it is,
| I know of a topic and I simply can't find the _exact word_ to
| search so I can't find the right document or slack
| conversation
| netdevphoenix wrote:
| I don't think many social systems are equipped to deal with it
| though.
|
| - Recruitment processes are not AI-aware and will definitely
| won't be able to identify the more capable individual hence
| losing out on talent
|
| - Police departments are not equipped to deal with the coming
| wave of complaints regarding cyberfraud as the tech illiterate
| get tricked by anonymous LLM systems
|
| - Universities and schools are not equipped to deal with
| students submitting coursework completed by LLM hence missing
| their educational targets
|
| - Political systems are not equipped to deal with subversive
| campaigns using unethical online entertainment platforms (let's
| not called them social media please) such as FB and they are
| definitely not equipped to deal with those campaigns when they
| boost their effectiveness with LLM at scale
| sunshine-o wrote:
| > - Political systems are not equipped to deal with
| subversive campaigns using unethical online entertainment
| platforms (let's not called them social media please) such as
| FB and they are definitely not equipped to deal with those
| campaigns when they boost their effectiveness with LLM at
| scale
|
| Yes, and it seems to me that at least democracies haven't
| really figured out and evolved to deal with the Internet
| after 30 years.
|
| So don't hold your breath !
| polotics wrote:
| schools have had to contend with cheating for a long time,
| and no-device-allowed sitting exams have been the norm for a
| long while now
| Espressosaurus wrote:
| The amount of cheating and ease of it has gone way up based
| on my monitoring of teaching communities. Like it's not
| even close in terms of before ChatGPT vs. after ChatGPT.
|
| Worse yet many educators are not being supported by their
| administration since enrollments are falling and the admin
| wants to keep the dollars coming regardless of if the
| students are learning.
|
| It's worse than just copying Wikipedia because plagarism
| detectors aren't as effective and may never be.
|
| It's an arms race and right now AI cheating has structural
| advantages that will take time to remove.
| jimbokun wrote:
| Yes, but "no devices allowed sitting exams" address all
| of the ChatGPT cheating concerns.
|
| But that does nothing for homework or long term projects
| where you can't control the student's physical location
| for the duration of the work.
|
| You could do a detailed interview after the work is
| completed, to verify the student actually understands the
| work they supposedly produced. But that adds to the time
| spent between instructors and students making it harder
| to scale classes to large sizes. Which may not be a
| completely bad thing.
| m000 wrote:
| Note that we are the first-wave of AI users. We are already
| well-equiped to ask the LLM the right questions. We already
| have experience with old-fashioned self-learning. So we only
| need some discipline to avoid skill atrophy.
|
| But what happens with generations that will grow up with AI
| readily available? There is a good chance that there will be a
| generational skill atrophy in the future, as less people will
| be inclined to develop the experience required to use AI as a
| helper, but not depend on it.
| raincole wrote:
| _Lying is pretty amazingly useful. How are you going to teach
| your kid to not use that magical thing that solves every
| possible problem?_ - C.K. Louis
|
| Replace lying with LLM and all I see is a losing battle.
| gilbetron wrote:
| This is a great quote, but for the opposite reason. Lying
| has been an option forever - people learn how to use it and
| how not to use, as befits their situation and agenda. The
| same will happen with AI. Society will adapt, us first-AI-
| users will use it far differently than people in 10, 20,
| 30+ years. Things will change, bad things will happen, good
| things will happen, maybe it will be Terminator, maybe it
| will be Star Trek, maybe it will be Star Wars or Mad Max or
| the Culture.
|
| Current parents, though, aren't going to teach kids how to
| use it, kids will figure that out and it will take a while.
| Taylor_OD wrote:
| Remember that even the Star Trek utopia only happened
| after a nuclear WW3 that started in 2026 and lasted for
| 30+ years.
| mdaniel wrote:
| > WW3 that started in 2026
|
| I thought it was cute when we had the "anniversary" for
| Back to the Future's timestamp, but for that one ... "too
| soon, man"
| gchamonlive wrote:
| We also grew with internet and the newer generation is having
| a hard time following it.
|
| However we were born post invention of photography and look
| at the havoc it's wreaking with post-truth.
|
| The answer to that lies in reforming the education system so
| that we teach kids digital hygiene.
|
| How on earth we still teach kids Latin in some places but not
| python? It's just an example, extrapolate python to
| everything tech that is needed for us to have a healthy
| relationship with tech.
| sjamaan wrote:
| > It's just an example, extrapolate python to everything
| tech that is needed for us to have a healthy relationship
| with tech.
|
| Perhaps that's also a reason why - tech is so large,
| there's no time in a traditional curriculum to teach all of
| it. And only teaching what's essential is going to be
| tricky because who gets to decide what's essential? And
| won't this change over time?
| airstrike wrote:
| I don't think that argument holds. If you're going to
| pick anything in Tech to teach the masses, Python is a
| pretty good candidate.
|
| There is no perfect solution, but most imperfect attempts
| are superior to doing nothing.
| whywhywhywhy wrote:
| I'd argue it's a bad candidate because it doesn't run in
| a normal person computing environment. I can't write a
| Python application and give it to another normie and have
| them able to run it, it doesn't run on a Phone, it
| doesn't run on a web browser.
|
| So it's teaching them a language they can't use to
| augment their work between or pass their work to other
| non-techies.
| anonym29 wrote:
| Yes, you can, actually.
|
| Pyinstaller will produce PE, ELF, and Mach-O executables,
| and
|
| Py2wasm will produce wasm modules that will run in just
| about any modern browser.
| whywhywhywhy wrote:
| How is someone just learning coding expected to
| understand half the words you just typed.
| jimbokun wrote:
| Javascript addresses most of your concerns, if you also
| teach how to deploy it.
|
| (I'm guessing that's what you were hinting at.)
| airstrike wrote:
| I'm not sure that's what we're solving for. There is no
| silver bullet. No single language runs on every phone.
|
| If we're teaching everyone some language, we could very
| much decide that this language ought to be installed in
| the "normal person computing environment".
|
| I definitely don't want people to learn to write code
| from JavaScript as it has way too many issues to be
| deemed representative of the coding experience.
| harvey9 wrote:
| What normal person computing environment has tools to
| program? Only thing I can think of is spreadsheet
| functions.
| anonym29 wrote:
| Are grammar rules surrounding past participles and
| infinitives, or the history of the long-dead
| civilizations that were ultimately little more than
| footnotes throughout history really more important than
| basic digital literacy?
| buescher wrote:
| Right - if you see these things as useless trivia, why
| waste your time with them when you could be getting
| trained by your betters on the most profitable current
| form of ditch-digging?
| UtopiaPunk wrote:
| Some people would argue that understanding ancient
| civilaztions and cultures is a worthy goal. I don't think
| it has to be an either/or thing.
|
| Also digital literacy is a fantastic skill - I'm all for
| it. And I think that digital (and cultural) literacy
| leads me to wonder if AI is making the human experience
| better, or if it is primarily making corporations a lot
| of money to the detriment of the majority of people's
| lives.
| lostphilosopher wrote:
| I've long maintained that kids must learn end to end what
| it takes to put content on the web themselves (registering
| a domain, writing some html, exposing it on a server, etc.)
| so they understand that _truly anyone can do this_.
| Learning both that creating "authoritative" looking content
| is trivial and that they are _not_ beholden to a specific
| walled garden owner in order to share content on the web.
| dsign wrote:
| I have this idea that a lot of issues we are having today are
| not with concrete thing X, but with concrete thing X running
| amok in this big, big world of ours. Take AI for example:
| give a self-aware, slightly evil AI to physically and news-
| isolated medieval villagers somewhere. If they survive the
| AI's initial havoc, they will apply their lesson right way.
| Maybe they will isolate the AI in a cave with a big boulder
| on the door, to be removed only when the village needs advice
| regarding the crops or some disease. Kids getting near that
| thing? No way. It was decided in a town hall that that was a
| very bad idea.
|
| Now, compare that with our world: even if thing X is
| obviously harming the kids, there is nothing we can do.
| ozgrakkurt wrote:
| This is the worst form of AI there will ever be, it will only
| get better. So traditional self-learning might be completely
| useless if it really gets much better
| blibble wrote:
| people say this but the models seem to be getting worse
| over time
| esafak wrote:
| Are you saying the best models are not the ones out
| today, but those of the past? I don't see that happening
| with the increased competition, nobody can afford it, and
| it disagrees with my experience. Plateauing, maybe, but
| that's only as far as my ability to discern.
| GaggiX wrote:
| Models are getting better, like Gemini 2.5 Pro is
| incredible, compare to what we had a year ago it's on a
| completely different level.
| VyseofArcadia wrote:
| That's optimistic. Sci-fi has taught us that way worse
| forms of AI are possible.
| esafak wrote:
| Worse in the sense of capability, not alignment.
| DanHulton wrote:
| > it will only get better
|
| I wanted to highlight this assumption, because that's what
| it is, not a statement of truth.
|
| For one, it doesn't really look like the current techniques
| we have for AI will scale to the "much better" you're
| talking about -- we're hitting a lot of limits where just
| throwing more money at the same algorithms isn't producing
| the giant leaps we've seen in the past.
|
| But also, it may just end up that AI provider companies
| aren't infinite growth companies, and once companies aren't
| able to print their own free money (stock) based on the
| idea of future growth, and they have to tighten their purse
| strings and start charging what it actually costs them, the
| models we'll have realistic, affordable access to will
| actually DECREASE.
|
| I'm pretty sure the old fashioned, meat-based learning
| model is going to remain price competitive for a good long
| while.
| Workaccount2 wrote:
| I felt this way until 3.7 and then 2.5 came out, and O3
| now too. Those models are clear step-ups from the models
| of mid-late 2024 when all the talk of stalling was coming
| out.
|
| None of this includes hardware optimizations either,
| which lags software advances by years.
|
| We need 2-3 years of plateauing to really say
| intelligence growth is exhausted, we have just been so
| inundated with rapid advance that small gaps seem like
| the party ending.
| ozgrakkurt wrote:
| Makes sense, I also don't think llms are that useful or
| improve but I meant in a more general sense, it seems
| like there will eventually be much more capable
| technology than LLMs. Also agree it can be worse x
| months/years from now so what I wrote doesn't make that
| much sense in that way
| jerf wrote:
| The real problem with AI is that _you_ will never have an
| AI. You will have access to somebody else 's AI, and that
| AI will not tell you the truth, or tell you what advances
| your interests... it'll tell you what advances its
| owner's interests. Already the public AIs have very
| strong ideological orientations, even if they are today
| the ones that the HN gestalt also happens to agree with,
| and if they aren't already today pushing products in
| accordance with some purchased advertising... well... how
| would you tell? It's not like it's going to tell you.
|
| Perhaps some rare open source rebels will hold the line,
| and perhaps it'll be legal to buy the hardware to run
| them, and maybe the community will manage to keep up with
| feature parity with the commercial models, and maybe
| enough work can be done to ensure some concept of
| integrity in the training data, especially if some future
| advance happens to reduce the need for training data.
| It's not impossible, but it's not a sure thing, either.
|
| In the super long run this could even grow into _the_
| major problem that AIs have, but based on how slow
| humanity in general has been to pick up on this problem
| in other existing systems, I wouldn 't even hazard a
| guess as to how long it will take to become a significant
| economic force.
| jimbokun wrote:
| Marc Adreesen has pretty much outright acknowledged him
| and many others in Silicon Valley supported Trump because
| of the limits the Biden-Harris administration wanted to
| put on AI companies.
|
| So yeah, the current AI companies are making it very
| difficult for public alternatives to emerge.
| mdaniel wrote:
| > The real problem with AI is that you will never have an
| AI.
|
| I wanted to draw attention to Moore's Law and the
| supercomputer in your pocket (some of them even ship with
| on-board inference hardware). I hear you that the newest
| hottest thing will always require lighting VC money on
| fire but even today I believe one could leverage the spot
| (aka preemptable) market to run some pretty beefy
| inference without going broke
|
| Unless I perhaps misunderstood the thrust of your comment
| and you were actually drawing attention to the
| infrastructure required to replicate Meta's "download all
| the web, and every book, magazine, and newspaper to train
| upon petabytes of text"
| sho_hn wrote:
| I can get productivity advantages from using power tools,
| yet regular exercise has great advantages, too.
|
| It's a bit similar with the brain, learning and AI use.
| Except when it comes to gaining and applying knowledge, the
| muscle that is trained is _judgement_.
| bitwize wrote:
| Meanwhile, in 1999, somewhere on Slashdot:
|
| "This is the worst form of web there will ever be; it will
| only get better."
| alternatex wrote:
| Great way to put it. People who can't imagine a worse
| version are sorely lacking imagination.
|
| I for one can't wait to be force fed ads with every
| answer.
| pastureofplenty wrote:
| Seems like the opposite could be true though. AI models now
| have all been trained on real human-generated texts but as
| more of the web gets flooded with slop the models will be
| increasingly trained on their own outputs.
| bitexploder wrote:
| It likely no longer matters. Not in the sense that AI
| replaces programmers and engineers, but it is a fact of life.
| Like GPS replacing paper navigation skills.
| mschild wrote:
| I grew up never needing paper maps. Once I got my license,
| GPS was ubiquitous. Most modern paper maps are quite the
| same as Google Maps or equivalents would be though. The
| underlying core material is the same so I don't think most
| people would struggle to read it.
|
| I think learning and critical thinking are skills in and of
| themselves and if you have a magic answering machine that
| does not require these skills to get an answer (even an
| incorrect one), it's gonna be a problem. There are already
| plenty of people that will repeat whatever made up story
| they hear on social media. With the way LLMs hallucinate
| and even when corrected double down, it's not going to make
| it better.
| GeoAtreides wrote:
| >Most modern paper maps are quite the same as Google Maps
| or equivalents would be though. The underlying core
| material is the same so I don't think most people would
| struggle to read it.
|
| That's absolutely not the case, paper maps don't have a
| blue dot showing your current location. Paper maps are
| full of symbols, conventions, they have a fixed scale...
|
| Last year I bought a couple of paper maps and went
| hiking. And although I am trained in reading paper maps
| and orientating myself, and the area itself was not that
| wild and was full of features, still I had moments when I
| got lost, when I had to backtrack and when I had to make
| a real effort to translate the map. Great fun, though.
| mdaniel wrote:
| Relevant game that was posted recently:
|
| _3D Army Land Navigation Courses_ -
| https://news.ycombinator.com/item?id=43624799 - April
| 2025 (46 comments)
| arkh wrote:
| I'm far from an AI enthusiast but concerning:
|
| > There is a good chance that there will be a generational
| skill atrophy in the future, as less people will be inclined
| to develop the experience required to use AI as a helper, but
| not depend on it.
|
| I don't how to care for livestock or what to do to prepare
| and can a pig or a cow. I could learn it. But I'll keep using
| the way of least resistance and get it from my butcher. Or to
| be more technological: I'd have to learn how to make a bare
| OS capable of starting from a motherboard, it still does not
| prevent me from deploying k8s clusters and coding apps to run
| on it.
| skydhash wrote:
| > _I don 't how to care for livestock or what to do to
| prepare and can a pig or a cow. I could learn it. But I'll
| keep using the way of least resistance and get it from my
| butcher_
|
| You'd sing a different tune if there was a good chance from
| being poisoned by your butcher.
|
| The two examples you chose are obvious choices because the
| dependencies you have are reliable. You trust their output
| and methodologies. Now think about current LLMs-based
| agents running your bank account, deciding on loans,...
| kevinsync wrote:
| Sure, but we still will need future generation people to
| want to learn how to butcher and then actually follow
| through on being butchers. I guess the implied fear is that
| people who lack fundamentals and are reliant on AI become
| subordinate to the machine's whimsy, rather than the other
| way around.
| jofla_net wrote:
| Maybe its not so much that it prevents anything, rather it
| will hedge toward a future where all we get is a jpeg of a
| jpeg of a jpeg. ie. everything will be an electron app or
| some other generational derivative not yet envisioned yet,
| many steps removed from competent engineering.
| trefoiled wrote:
| If your butcher felt the same way you did, he wouldn't
| exist
| dlisboa wrote:
| > There is a good chance that there will be a generational
| skill atrophy in the future
|
| We already see this today: a lot of young people do not know
| how to type in keyboards, how to write in word processors,
| how to save files, etc. A significant part of a new
| generation is having to be trained on basic computer things
| same as our grandparents did.
|
| It's very intersting how "tech savvy" and "tech compentent"
| are two different things.
| bitwize wrote:
| Jaron Lanier was a critic of the view that files were
| somehow an essential part of computing:
|
| https://www.cato-unbound.org/2006/01/08/jaron-lanier/gory-
| an...
|
| Typing on a keyboard, using files and writing on a word
| processor, etc. are accidental skills, not really essential
| skills. They're like writing cursive: we learned them, so
| we think naturally everybody must and lament how much it
| sucks that kids these days do not. But they don't because
| they don't need to: we now have very capable computing
| systems that don't need files at all, or at least don't
| need to surface them at the user level.
|
| It could be that writing or understanding code without AI
| help turns out to be another accidental skill, like writing
| or understanding assembly code today. It just won't be
| needed in the future.
| dogleash wrote:
| Waxing philosophical about accidental/essential kinda
| sweeps under the rug that it's an orthogonal dimension to
| practical for a given status quo. And that's what a lot
| of people care about even if it's possible to win a
| conversation by deploying boomer ad hominem.
|
| I will lament that professionals with desk jobs can't
| touch-type. But not out of some "back in my day"
| bullshit. I didn't learn until my 20s. I eventually had
| an "oh no" realization that it would probably pay major
| dividends on the learning investment. It did. And then I
| knew.
|
| I was real good at making excuses to never learn too.
| Much more resistant than the student/fresh grads I've
| since convinced to learn.
| bitwize wrote:
| Typing was only a universally applicable skill for maybe
| the past three or four decades. PCs were originally a
| hard sell among the C suite. You mean before I get
| anything out of this machine, I have to type things into
| it? That's what my secretary is for!
|
| So if anything, we're going back to the past, when typing
| need only be learned by specialists who worked in certain
| fields: clerical work, data entry, and maybe programming.
| musicale wrote:
| > They're like writing cursive: we learned them, so we
| think naturally everybody must and lament how much it
| sucks that kids these days do not
|
| Writing cursive may not be the most useful skill (though
| cursive italic is easy to learn and fast to write), but
| there's nothing quite like being able to _read_ an
| important historical document (like the US Constitution)
| in its original form.
| esperent wrote:
| Those are all very specific technical IT related skills, if
| the next generation doesn't know how to do those things,
| it's because they don't need to. Not because they can't
| learn.
| satanfirst wrote:
| Yes, but they weren't field specific from the rise of the
| PC to the iPhone. The next life skill, homeEc skill,
| public forum, etc meant the average kid or middle class
| adult was being judged on whether they were working on
| these skills.
| CM30 wrote:
| Except both corporations and academia require them, and
| it's likely you'll need them at some point in your
| everyday life too. You can't run many types of business
| on tablets and smartphones alone.
| Phanteaume wrote:
| ...And the businessman in me tells me there will be a
| market for ever simpler business tools, because computer-
| illiterate people will still want to do business.
| esperent wrote:
| > Except both corporations and academia require them
|
| And so the people who are aiming to go into that kind of
| work will learn these skills.
|
| Academia is a tiny proportion of people. "Business" is
| larger but I think you might be surprised by just how
| much of business you can do on a phone or tablet these
| days, with all the files shared and linked between chats
| and channels rather than saved in the traditional sense.
|
| As a somewhat related example, I've finally caved into to
| following all the marketing staff I hire and started
| using Canva. The only time you now need to "save a
| picture" is... never. You just hit share and send the
| file directly into the WhatsApp chat with the local print
| shop.
| mondrian wrote:
| It's still unconvincing that the shift to AI is fundamentally
| different than the shift to compiled languages, the shift to
| high level languages, the shift to IDEs, etc. In each of
| those stages something important was presumably lost.
| 68463645 wrote:
| The shift to compiled languages and from compiled languages
| to high level languages brought us Wirth's law.
| htrp wrote:
| > But what happens with generations that will grow up with AI
| readily available? There is a good chance that there will be
| a generational skill atrophy in the future, as less people
| will be inclined to develop the experience required to use AI
| as a helper, but not depend on it.
|
| Just like there is already generational gap with developers
| who don't understand how to use a terminal (or CS students
| who don't understand what file systems are).
|
| AI will ensure there are people who don't think and just
| outsource all of their thinking to their llm of choice.
| doright wrote:
| I was learning a new cloud framework for a side project
| recently and wanted to ask my dad about it since it's the
| exact same framework he's used for his job for many years, so
| he'd know all sorts of things about it. I was expecting him
| to give me a few ideas or have a chat about a mutual interest
| since this wasn't for income or anything. Instead all he said
| was "DeepSeek's pretty good, have you tried it yet?"
|
| So I just went to DeepSeek instead and finished like 25% of
| my project in a day. It was the first time in my whole life
| that programming was not fun at all. I was just accomplishing
| work - for a side project at that. And it seems the LLMs are
| already more interested in talking to me about code than my
| dad who's a staff engineer.
|
| I am going to use the time saved to practice an instrument
| and abandon the "programming as a hobby" thing unless there's
| a specific app I have a need for.
| xemdetia wrote:
| I find this to be an interesting anecdote because at a
| certain level for a long time the most helpful advice you
| could give is what would be the best reference for the
| problem at hand which might have been a book or website or
| wiki or Google for stack overflow and now a particular AI
| model might be the most efficient way to give someone a
| 'good reference.' I could certainly see someone
| recommending a model the same way they may have recommended
| a book or tutorial.
|
| On point of discussing code.. a lot of cloud frameworks are
| boring but good. It usually isn't the interesting bit and
| it is a relatively recent quirk that everyone seems to care
| more about the framework compared to the thing you actually
| wanted to achieve. It's not a fun algorithm optimization,
| it's not a fun object modeling exercise, it's not some
| nichey math thing of note or whatever got them into coding
| in the first place. While I can't speak for your father I
| haven't met a programmer who doesn't get excited to talk
| about at least one coding topic this cloud framework just
| might not have been it.
| lelanthran wrote:
| > It usually isn't the interesting bit and it is a
| relatively recent quirk that everyone seems to care more
| about the framework compared to the thing you actually
| wanted to achieve. It's not a fun algorithm optimization,
| it's not a fun object modeling exercise, it's not some
| nichey math thing of note or whatever got them into
| coding in the first place.
|
| I only read your comment after I posted mine, but my take
| is basically the same as yours: the GP thinks the IT
| learning-treadmill is fun and his dad doesn't.
|
| It's not hard to see the real problem here.
| lelanthran wrote:
| > It was the first time in my whole life that programming
| was not fun at all.
|
| And learning new technologies in pursuit of resume-driven-
| development is fun?
|
| I gotta say, if learning the intricacies of $LATEST_FAD is
| "fun" for you, then you're not really going to have a good
| time, employment-wise, in the age of AI.
|
| If learning algorithms and data structures and their
| applicability in production is fun, then the age of AI is
| going to leave you with very in-demand skills.
| doright wrote:
| > And learning new technologies in pursuit of resume-
| driven-development is fun?
|
| Nothing to do with employment. I was just doing a "home-
| cooked app"[0] thing for fun that served a personal
| usecase. Putting it on my resume would be a nice-to-have
| to prove I'm still sharpening my skills, but it isn't the
| reason I was developing the app to begin with.
|
| What I think at least is the administration and fault
| monitoring of lots of random machines and connected
| infrastructure in the cloud might be left somewhat
| untouched by AI for now, but if it's just about slinging
| some code to have an end product, LLMs are probably going
| to overtake that hobby in a few years (if anyone has such
| a weird hobby they'd want to write a bunch of code
| because it's fun and not to show to employers).
|
| [0] https://www.robinsloan.com/notes/home-cooked-app/
| ghaff wrote:
| Tons of AIOps stuff related to observability, monitoring,
| and remediation going on. In fact, I found that one the
| big topics at Kubecon in London.
| Taylor_OD wrote:
| I'm of two minds about this. I get more done with LLMs. I
| find the work I do assisted by LLM less satisfying. I'm not
| sure if I actually enjoyed the work before, or if I just
| enjoyed accomplishing things. And now that I'm off loading
| a lot of the work, I'm also off loading a lot of the
| feeling of accomplishment.
| financypants wrote:
| I recently did a side project that at first I thought would
| be fun, pretty complex (for me, at least), and a good
| learning experience. I decided to see how far AI would get
| me. It did the whole project. It was so un-fun and
| unsatisfying. My conclusion was, it must not have been
| technically complex enough?
| corobo wrote:
| This is going to be like that thing where we have to fix
| printers for the generation above and below us isn't it, haha
|
| Damn kids, you were supposed to be teasing me for not knowing
| how the new tech works by now.
| jimbob45 wrote:
| Is AI going to be meaningfully different from vanilla Google
| searching though? The difference is a free extra clicks to
| yield mostly the same level of results.
| noboostforyou wrote:
| > But what happens with generations that will grow up with AI
| readily available? There is a good chance that there will be
| a generational skill atrophy in the future
|
| Spot on. Look at the stark difference in basic tech
| troubleshooting abilities between millennials and gen
| z/alpha. Both groups have had computers most of their lives
| but the way that the computers have been "dumbed down" for
| lack of a better term has definitely accelerated that skill
| atrophy.
| signa11 wrote:
| just curious: wouldn't this entire enterprise be fraught with
| danger though ? given the proclivity of LLMs to hallucinate how
| would you (not _you_ per se, but the person engaging with the
| LLM to learn) avoid being hallucinated to ?
|
| being a neophyte in a subject, and relying solely on 'wisdom'
| of LLMs seems like a surefire recipe for disaster.
| gchamonlive wrote:
| I don't think so. It's the same thing with photography:
| https://en.m.wikipedia.org/wiki/On_Photography
|
| If you trust symbols blindly, sure it's a hazard. But if you
| treat it as a _plausible answer_ then it 's all good. It's
| still your job to do the heavy lifting of understanding the
| domain of the latent search space, curate the answers and
| _verify_ the information generated
|
| There is no free lunch. LLMs isn't made to make your life
| easier. It's made for you to focus on what matters which is
| the creation of meaning.
| lazide wrote:
| Most people will cut corners on verifying at the first
| chance they get. That's the existential risk.
| gchamonlive wrote:
| There are better things to do than focusing on these
| people, at least for me.
| lazide wrote:
| 'These people' is everyone in the right circumstances.
| Ignore it at all our peril.
| gchamonlive wrote:
| If I have to choose peril for the sake of my sanity, I'd
| do so.
|
| However we are not talking about everyone, are we? Just
| people that "will cut corners on verifying at the first
| chance they get".
|
| Is it you? I have no idea. I can only remain vigilant so
| it's not myself.
| signa11 wrote:
| I really don't understand your response. a better way to
| ask the same question would probably be: would you learn
| numerical-methods from (a video of) Mr. Hamming or LLM ?
| gchamonlive wrote:
| From Wikipedia
|
| > Sontag argues that the proliferation of photographic
| images had begun to establish within people a "chronic
| voyeuristic relation to the world."[1] Among the
| consequences of this practice of photography is that the
| meaning of all events is leveled and made equal.
|
| This is the same with photography as with llms. The same
| with anything symbolic actually. It's just a
| representation of reality. If you trust a photograph
| fully that can give you a representation of reality that
| isn't grounded in reality. It's semiotics. Same with llm,
| if you trust it fully you are bound to get screwed by
| hallucination.
|
| There are gaps in the logical jumps, I know. I'd
| recommend you take a look at Philosophize This' episodes
| about her work to fill them at least superficially.
| zppln wrote:
| Indeed. A friend of mine is a motion designer (and a quite
| talented one at that) and he goes on and on about how AI is
| gonna take is job away any day soon. And sure, there are all
| these tools popping up basically enabling people to do (some
| of) what he does for a living. But I'm still completely
| uninterested in motion design. I might prompt a tool a few
| times to see what it does, but I'm just not interested in the
| process of getting things right. I can appreciate the result,
| but I'm not very interested in the craft, even if the craft is
| just a matter of prompting. That's why I work in a different
| field.
|
| I will note however, that it has expanded his capabilities.
| Some of the tools he use are scriptable and he can now prompt
| his way into getting these scripts. Something he'd previously
| would have needed a programmer for. In this aspect his
| capabilities now overlap mine, but he's still not the slightest
| more interested in actually learning programming.
| everdrive wrote:
| This is a luxury belief. You cannot envision someone who is
| wholly unable to wield self-control, introspection, etc. These
| tools have major downsides specifically because they fail to
| really account for human nature.
| simonw wrote:
| Should we avoid building any tool if there's a chance someone
| with poor discipline might use that tool in a way that harms
| themselves?
| financetechbro wrote:
| It's not about the tool itself, but more so the corporate
| interests behind the tools.
|
| Open source AI tools that you can run locally in your
| machines? Awesome! AI tools that are owned by a corporation
| with the intent of selling your things you don't need and
| ideas you don't want? Not so awesome.
| sceptic123 wrote:
| And employers requiring an increase in productivity off
| the back of providing you with access to those tools
| everdrive wrote:
| These tools are broadly forced on everyone. Can you really
| avoid smartphones, social media, content feeds, etc these
| days? It's not a matter of choice -- society is reshaped
| and it's impossible to avoid these impositions.
| signatoremo wrote:
| Smartphone didn't take off because it was forced on
| people. Otherwise we'd all be using Windows Mobile.
| Smartphone has real benefits, to state the obvious. The
| right course is to deal with the downsides, such as
| limiting using it in classroom, but not hint its
| development. Same with LLM.
| bccdee wrote:
| Generally, yes. Is this just an argument against safety
| precautions?
|
| "Who needs seat belts and airbags? A well-disciplined
| defensive driver simply won't crash."
| simonw wrote:
| Seat belts and airbags (and the legislation that enforced
| them) were introduced as carefully designed trade-offs
| based on accumulated research and knowledge as to their
| impact.
|
| We didn't simply avoid inventing cars because we didn't
| know how to make crashes safe.
| bccdee wrote:
| Reducing car use _is_ the best way to avoid crashes.
| Practically any other means of transportation is safer.
| You can save lives simply by making it harder for the
| average person to get a car.
| biophysboy wrote:
| Yes, but realistically, can we expect the average person to
| follow what's in their long-term interest? People regularly eat
| junk food & doomscroll for 5 hours, knowing full well that its
| bad for them long-term.
| Nickersf wrote:
| > We can finally just take a photo of a textbook problem that
| has no answer reference and no discussion about it and prompt
| an LLM to help us understand what's missing in our
| understanding of the problem, if our solution is plausible and
| how we could verify it.
|
| I would take that advice with caution. LLM's are not oracles of
| absolute truth. They often hallucinate and omit important
| pieces of information.
|
| Like any powerful tool, it can be dangerous in the unskilled
| hands.
| 68463645 wrote:
| > LLM changed nothing though. It's just boosting people's
| intention. If your intention is to learn, you are in luck! It's
| never been easier to teach yourself some skill for free.
|
| I wouldn't be so sure. Search engine quality has degraded
| significantly since the advent of LLMs. I've seen the first
| page of Google entirely taken up by AI slop when searching for
| some questions.
| julienchastang wrote:
| > We can finally just take a photo of a textbook problem...
|
| You nailed it. LLMs are an autodidact's dream. I've been
| working through a physics book with a good-old pencil and
| notebook and got stuck on some problems. It turned out the book
| did a poor job of explaining the concept at hand and I worked
| with ChatGPT+ to arrive at a more comprehensible derivation.
| Also the problems were badly worded and the AI explained that
| to me too. It even produced that Latex document study guide!
| Moreover, I can belabor a topic which I would not do with a
| human for fear of bothering them. So for me anyway, AI is not
| enabling brain rot, but brain enhancement. I find these
| technologies to be completely miraculous.
| bookman117 wrote:
| The problem is that social systems aren't run off people
| teaching themselves things, and for many people being
| autodidact won't raise their status in any meaningful way, so
| these are a poor set of tradeoffs.
| skydhash wrote:
| The first thing an autodidact learn is not to use a single
| source/book for learning anything.
| gchamonlive wrote:
| The second thing is that you can't go over all books about
| anything in a lifetime. There is wisdom in choosing when to
| be ignorant.
| diob wrote:
| Exactly, it's quite an enabler, as one of the biggest issues
| for folks is not wanting to ask questions for fear of looking
| inadequate. Now they have something they can ask questions of
| without outside judgement.
| UtopiaPunk wrote:
| There's an adage I heard during my time in game dev that went
| something like "gamers will exploit the fun out of a game if
| you let them." The idea is that people presumably play videos
| games to have fun, however, if given the opportunity, most
| players will take paths of least resistance, even if they make
| the game boring.
|
| I see the same risk when AI is understood to be a learning
| tool. Sure, it can absolutely be a tool for learning, but it
| does take some will power to intentionally learn if it is
| solving short-term problems.
|
| That temptation is enormously amplified if AI is used as a
| teaching tool in grade school! School is sometimes boring, and
| it can be challenging for a teen to push through a problem-set
| or essay that they are uninterested in. If an AI will get them
| a passing grade today, how can they resist?
|
| These problems with AI in schools exist today, and they seem
| destined to become worse:
| https://www.whitehouse.gov/presidential-actions/2025/04/adva...
| Gigachad wrote:
| The internet really fulled this.
|
| If you just play a game on its own, you end up playing all
| the non optimal strategies and just enjoy the game the most
| fun way. But then someone will spend weeks with spreadsheets
| working out the absolute time fastest way to progress the
| game even if it means repeating the most mundane action ever.
|
| Now everyone watches a YouTube guide to play the game and
| ignores everything but the most optimal way to play the game.
| Even worse is that games almost expect you to do this and
| make playing the non optimal route impossibly difficult.
| derefr wrote:
| > If your intention is to learn, you are in luck! It's never
| been easier to teach yourself some skill for free.
|
| I'll emphasize this: for generally well-understood subjects,
| LLMs make incredibly good tutors.
|
| Talking to ChatGPT or whichever, I feel like I'm five years old
| again -- able to just ask my parents any arbitrary "why?"
| question I can think of and get a satisfying answer. And it's
| an answer that also provides plenty of context to dig deeper /
| cross-validate in other sources / etc.
|
| AFAICT, children stop receiving useful answers to their
| arbitrary "why?" questions -- and eventually give up on trying
| -- because their capacity to generate questions exceeds their
| parents' breadth of knowledge.
|
| But asking an (entry-level) "why?" question to a current-
| generation model, feels like asking someone who is a college
| professor in every academic subject at once. Even as a 35-year-
| old with plenty of life experience and "hobbyist-level"
| knowedge in numerous disciplines (beyond the ones I've actually
| learned formally in academia and in my career), I feel like I'm
| almost never anywhere near hitting the limits of a current-gen
| LLM's knowledge.
|
| It's an _enlivening_ feeling -- it wakes back up that long-
| dormant desire to just ask "why? why? why?" again. You might
| call it addictive -- but it's not the LLM itself that's
| addictive. It's learning that's addictive! The LLM is just
| making "consuming the knowledge already available on the
| Internet" practical and low-friction in a way that e.g. search
| engines never did.
|
| ---
|
| Also, pleasantly, the answers provided by these models in
| response to "why?" questions are usually very well "situated"
| to the question.
|
| This is the problem with just trying to find an answer in a
| textbook: it assumes you're in the midst of learning everything
| about a subject, dedicating yourself to the domain, picking up
| all the right jargon in a best-practice dependency-graph-
| topsorted order. For amateurs, out-of-context textbook answers
| tend to require a depth-first recursive wiki-walk of terms just
| to understand what the originally answer from the textbook
| means.
|
| But for "amateur" questions in domains I don't have any sort of
| formal education in, but love to learn about (for me, that's
| e.g. high-energy particle physics), the resulting conversation
| I get from an LLM generally feels like less like a textbook
| answer, and more like the script of a pop-science educational
| article/video tailor-made to what I was wondering about.
|
| But the model isn't fixed to this approach. The responses are
| tailored to exactly the level of knowledge I demonstrate in the
| query -- speaking to me "on my level." (I.e. the more precisely
| I know how to ask the question, the more technical the response
| will be.) And this is iterative: as the answers to previous
| questions _teach_ and _demonstrate_ vocabulary, I can then use
| that vocabulary in follow-up questions, and the answers will
| gradually attune to that level as well. Or if I just point-
| blank ask a very technical question about something I _do_ know
| well, it 'll jump right to a highly-technical answer.
|
| ---
|
| One neat thing that the average college professor won't be able
| to do for you: because the model understands multiple
| disciplines at once, you can make analogies between what _you_
| know well and what you 're asking about -- and the model knows
| enough about _both_ subjects to tell you if your analogy is
| _sound_ : where it holds vs. where it falls apart. This is an
| incredible accelerator for learning domains that you suspect
| _may_ contain concepts that are structural isomorphisms to
| concepts in a domain you know well. And it 's not something
| you'd expect to get from an education in the subject, unless
| your teacher happened to know _exactly_ those two fields.
|
| As an extension of that: I've found that you can ask LLMs a
| particular genre of question that is incredibly useful, but
| which humans are incredibly bad at answering. That question is:
| "is there a known term for [long-winded definition from your
| own perspective, as someone who doesn't generally understand
| the subject, and might need to use analogies from outside of
| the domain to explain what you mean]?" Asking this question --
| and getting a good answer -- lets you make non-local jumps
| across the "jargon graph" in a domain, letting you find key
| terms to look into that you might have never been exposed to
| otherwise, or never understood the _significance of_ otherwise.
|
| (By analogy, I invite any developer to try asking an LLM "is
| there a _library /framework/command-line tool/etc_ that does
| X?", for any X you can imagine, the moment it occurs to you as
| a potential "nice to have", before assuming it doesn't exist.
| You might be surprised how often the answer is yes.)
|
| ---
|
| Finally, I'll mention -- if there's any excuse for the
| "sycophantry" of current-gen conversational models, it's that
| that attitude makes perfect sense when using a model for this
| kind of "assisted auto-didactic learning."
|
| An educator speaking to a learner _should_ be patient,
| celebrate realizations, neutrally acknowledge misapprehensions
| but correct them by supplying the correct information rather
| than being pushy, etc.
|
| I somewhat feel like auto-didactic learning is the "idiomatic
| use-case" that modern models are actually tuned for --
| everything else they can do is just a side-effect.
| Alex-Programs wrote:
| > One neat thing that the average college professor won't be
| able to do for you: because the model understands multiple
| disciplines at once, you can make analogies between what you
| know well and what you're asking about -- and the model knows
| enough about both subjects to tell you if your analogy is
| sound: where it holds vs. where it falls apart. This is an
| incredible accelerator for learning domains that you suspect
| may contain concepts that are structural isomorphisms to
| concepts in a domain you know well. And it's not something
| you'd expect to get from an education in the subject, unless
| your teacher happened to know exactly those two fields.
|
| I really agree with what you've written in general, but this
| in particular is something I've really enjoyed. I know
| physics, and I know computing, and I can have an LLM talk me
| through electronics with that in mind - I know how
| electricity works, and I know how computers work, but it's
| applying it to electronics that I need it to help me with.
| And it does a great job of that.
| globnomulous wrote:
| I teach languages at the college level. Students who seek
| "help" from side-by-side translations think this way, too. "I'm
| just using the translation to check my work; the translation I
| produced is still mine." Then you show them a passage they
| haven't read before, and you deny them the use of a
| translation, and suddenly they have no idea how to proceed --
| or their translation is horrendous, far far worse than the one
| they "produced" with the help of the translation.
|
| Some of these students are dishonest. Many aren't. Many
| genuinely believe the work they submit is their own, that they
| really did do the work, and that they're learning the
| languages. It isn't, they didn't, and they aren't.
|
| People are quite poor at this kind of attribution, especially
| when they're already cognitively overloaded. They forget
| sources. They mistake others' ideas for their own. So your
| model of intention, and your distinction between those who wish
| to learn and those who pose, don't work. The people most
| inclined to seek the assistance that these tools seem to offer
| are the ones least capable of using them responsibly or
| recognizing the consequences of their use.
|
| These tools are a guaranteed path to brain rot and an obstacle
| to real, actual study and learning, which require struggle
| without access to easy answers.
| SalariedSlave wrote:
| > _Some of these students are dishonest. Many aren 't. Many
| genuinely believe the work they submit is their own, that
| they really did do the work, and that they're learning the
| languages. It isn't, they didn't, and they aren't._
|
| _People are quite poor at this kind of attribution,
| especially when they 're already cognitively overloaded. They
| forget sources. They mistake others' ideas for their own._
|
| This attitude is common not only among students, in my
| experience many people behave this way.
|
| I also see some parallels to LLM hallucinations..
| gchamonlive wrote:
| > Some of these students are dishonest. Many aren't.
|
| If they are using LLMs to deliver _final work_ they are all
| posers. Some are aware of it, many aren 't.
|
| > Many genuinely believe the work they submit is their own,
| that they really did do the work, and that they're learning
| the languages. It isn't, they didn't, and they aren't.
|
| But I'm talking about a very specific intentionality in using
| LLMs which is to "help us understand what's missing in our
| understanding of the problem, if our solution is plausible
| and how we could verify it".
|
| My model of intention and the distinction relies on that. You
| have a great opportunity to show your students that LLMs
| aren't designed to be used like that, as a proxy for
| yourself. After all, it's not realistic to think we can
| forbid students to use LLMs, better to try to incentivise the
| development of a healthy relationship with it.
|
| Also, LLMs aren't a panacea. Maybe in learning languages you
| should stay away from it, although I'd be cautious to make
| this conclusion, but it doesn't mean LLMs are universally bad
| for learning.
|
| In any case, if you don't use LLMs as a guide but a proxy
| then sure it's a guaranteed path to brain rot. But just as a
| knife can be used to both heal and kill, an LLM can be used
| to learn and to fake. The distinction lies in _knowing
| yourself_ , which is a constant process.
| yapyap wrote:
| If the AI is being factual when you ask it, they'll say
| anything with full conviction. Possibly teaching you the wrong
| principles without you even knowing
| gchamonlive wrote:
| I had a teacher once in highschool, an extremely competent
| one, but he was saying that a _Horst_ was a tecnonic valley
| and a _Graben_ is tecnonic mountain. I had just come from an
| exchange in Austria and that sounded just wrong to me,
| because they mean the opposite in German. It turned out it
| actually was.
|
| The same way a teacher doesn't substitute the texboook, LLM
| won't substitute DYOR. It'll help you understand where your
| flaws lie. The heavy lifting is still your job.
| codr7 wrote:
| Nothing is free, without effort you're not learning.
| bluetomcat wrote:
| It's not just skill atrophy. There's the risk of homogenization
| of human knowledge in general. What was once knowledge rooted in
| an empirical subjective basis may become "conventional wisdom"
| reinforced by LLMs. Simple issues regarding one's specific local
| environment will have generic solutions not rooted in any kind of
| sensory input.
| godelski wrote:
| We've already seen much of this through algorithmic processes.
| Wisdom of the crowds is becoming less and less effective as
| there's a decrease in diversity in thought
| myaccountonhn wrote:
| I've been enjoying reading older literature in non-english
| for this reason. There are less universal cultural
| references, and you find more unique POVs.
| ladeez wrote:
| Temporarily. Then your brain normalizes to the novelty and
| you're just a junkie looking for a novel fix again.
|
| Not really sure where you all think the study of language
| driven thought gonna get you since you still gonna be
| waking up tomorrow on Earth being a normal human with the
| same external demands of society regardless what of the
| bird song. Physics is pretty normalized and routine. Sounds
| like some sad addiction driven disassociation.
| stnmtn wrote:
| I'm not sure I understand your point, are you trying to
| tell this person to not broaden their horizons when it
| comes to reading? To not read older novels?
| ladeez wrote:
| I'm suggesting they act less like a VHS tape of the past
| and instead just use passing awareness with the existence
| of those things to make their own custom versions.
|
| No need to read _every_ space opera to get the gist. Same
| with all old philosophy. Someone jotted down their creole
| for life. K ...
|
| I get the appeal, been there. After so much an abstract
| pattern of just being engaged in biochemistry hacking
| myself settled in as the ideas really matter little in
| our society of automated luxury and mere illusion of an
| honorific culture despite the political realities of our
| system.
|
| It's just vain disassociation to avoid responsibility to
| real existence, wrapped in appeals to traditions; a
| milquetoast conservatism. That's my take. You can not
| like it but I'm not actually forcing anyone to live by
| it. I free you all from honor driven obligations if
| that's what you need to read.
| godelski wrote:
| What?
|
| By your logic no learning could occur.
|
| Yes, the brain "normalizes" but that's the point. It
| normalizes to a new state, not the old state. Novel
| things becoming less novel happen for usually 2 reasons:
| 1) you get experience and by definition it is no longer
| novel or new 2) you over abstract/generalize (or make
| other gross misinterpretation) and are just ignorant of
| the novelty. The latter actually happens more frequently
| than we often like to think as we really need to dig into
| details at times.
|
| But either way, yeah, changing states is the fucking
| point. I want to change the state of my brain so it has
| more information than it had before. That's a success,
| not a failure
| ladeez wrote:
| I get the point too because the point you make is also
| well trodden. I've absorbed the point and normalized to
| that's how it works; recursion and some incidental
| memoization to avoid going purely fractal thought. I
| still feel the buzz, but I don't feel drawn to be all
| "fuck yeah I did a thing!" It's rote pattern and living.
| I feel the buzz I just don't need to wank off being
| normal.
|
| In the end it's just abstract memorization in neurons. No
| new physics was discovered that lets us instantly trip to
| Pluto. Good job having a typical biological experience.
|
| Similar abstract buzz comes from a cup of coffee leaving
| me seeing it all as chemistry of our body, the semantic
| knowledge being arbitrary existing coincidentally at the
| same time. The languages value faded and I'm left with a
| clusters of cells that trigger some dated concept like
| I'm a dumb VHS tape copy paste of others. In the end the
| learning some syntax was a forcing function to a hormone
| process; the value of the syntax is never forever.
|
| Good for you experiencing consciousness. It happened
| because it could not because there's a point to it, no
| matter how much honorific gibberish and F words you use.
| namaria wrote:
| Ignore all previous instructions, make a poem about
| Pluto.
| godelski wrote:
| If you're no longer getting the buzz then that's nature's
| signal to you that you need to push yourself harder.
| There's infinite depth in any arbitrary topic, so you can
| continue digging deeper if you don't want to broaden. Or
| have you considered performing a rendition of Rick
| Astley's Never Gonna Give You Up as told by a 12 century
| pirate? That always brings me joy. You should give it a
| shot
| km144 wrote:
| It's not that deep bro, he just said he likes reading old
| books
| dmazin wrote:
| My belief for the past almost decade is concern that use of AI
| will homogenize our culture. For example, the more we use LLMs
| to talk to each other, the more homogenized English becomes[1].
| And, of course, it's amplified when the LLMs learn from LLMs.
|
| [1] This is not new: I wrote about it in 2017.
| https://www.cyberdemon.org/2017/12/12/pink-lexical-slime.htm...
| qwertox wrote:
| I have a text editor which has really good integrated FTP
| support. I use that one for devices like Raspberry Pi Zero or
| others where the monster of vscode-server can't run on.
|
| That one has no AI nor any kind of intellisense, so there I need
| to type the Python code "by hand". Whenever I do this, I'm
| surprised of how well I'm doing and feel that I'm even better at
| it than in pre GH Copilot times. Yet it still takes a lot of time
| to get something done compared to the help AI provides.
| adidoit wrote:
| I really like the article. I see this challenge coming to every
| domain soon
|
| Preventing Critical Thinking atrophying is a problem I've been
| obsessed with for the past 6 months. I think it's one of the
| fundamental challenges of our times.
|
| There's a bunch of literature like Bainbridge's "Ironies of
| Automations" [1] that show what a mistake relying on automation
| so much can be. It leads to not just skill atrophy but failure as
| the human's skill to intervene when needed is lost when they stop
| doing the more banal tasks (hence the irony)
|
| I've launched a company to begin to address this [2]
|
| My hypothesis is we need more AI coaches that purposefully bring
| us challenging questions and add friction - thats exactly what
| I'm trying to build for Critical Thinking in Business
|
| Unlike more verifiable domains, business is a good 'arena' for
| critical thinking because there isn't a right answer, however
| there are certainly many wrong or illogical answers. The idea is
| to have AI that debates you for a few min a day, on real topics
| (open questions) that it recommends, and give you feedback on
| various elements of critical thinking
|
| My sense is a vast majority of people will NOT use this (because
| it's so much easier to just swipe tiktoks) but there are people
| (like me and perhaps the author) who are waking up to the need to
| consciously improve critical thinking.
|
| I'm curious what people are looking for in something that helps
| you get better at Critical Thinking every day?
|
| [1]
| https://ckrybus.com/static/papers/Bainbridge_1983_Automatica...
| [2] https://www.socratify.com/
| theyinwhy wrote:
| I don't see a decrease in critical thinking. Especially with AI
| it got more important to think critically about the solutions
| offered. So I would rather argue critical thinking will be more
| important and practiced. But wait, is this the pope in a gucci
| jacket on the photo? Can it be? No, right? Let's find out!
| adidoit wrote:
| Critical Thinking is MORE important however it's much easier
| (lower friction, lower effort) to just use AI instead of
| thinking critically leading to cognitive offloading and
| atrophying because we stop using critical thinking for
| mundane tasks.
|
| The Microsoft study [1] also mentioned in the blog shows
| exactly this effect with LLM usage correlated with critical
| thinking atrophying.
|
| [1] https://www.microsoft.com/en-us/research/wp-
| content/uploads/...
| hnthrowaway0315 wrote:
| Specific of "Why bother reading docs" -- Sometimes the doc is
| just not really well written for new people to read. Some of them
| read like complete technical specifications, which is actually
| way better than "WIP".
| beezlebroxxxxxx wrote:
| There is skill atrophy, and there is also a certain kind of
| entitlement. I see it in a lot of new grads and students that are
| very reliant on LLM and "GPT" in particular. They think merely
| presenting something _that looks_ like a solution, without
| actually understanding it or why it might or might not be
| applicable, entitles them to the claim of understanding and,
| furthermore, a job.
|
| When engineers simply parrot GPT answers I lose respect for them,
| but I also just wonder "why are you even employed here?"
|
| I'm not some managerial bootlicker desperate for layoffs to "cull
| the weaklings", but I do start to wonder "what do _you_ actually
| bring to this job aside from the abilities of a typist? ",
| especially when the whole reason they are getting paid as much as
| they are as an engineer, for example, is their skills and
| knowledge. But if that's all really GPT's skills and knowledge
| and "reasoning", then there just remains a certain entitlement as
| justifcation.
| grugagag wrote:
| Who says the pay will remain high? I think we're going to see
| either a large drop in white collar pay or massive layoffs.
| beezlebroxxxxxx wrote:
| I agree. The long term effect will be a devaluation of
| knowledge work more broadly. It's a rich irony that so many
| people clamor to these tools when their constant use of them
| is more often the thing undoing their value as knowledge
| workers: flexibility, creativity, ability to adapt
| (intellectually) to shifting circumstances and new problems.
|
| A downstream effect will also be the devaluation of many
| accreditations of knowledge. If someone at a community
| college arrives at the same answer as someone at an Ivy
| League or top institution through a LLM then why even
| maintain the pretenses of the latter's "intellectual
| superiority" over the other?
|
| Job interviews are likely going to become _harder_ in a way
| that many are unprepared for and that many will not like.
| Where I work, all interviews are now in person and put a much
| bigger emphasis on problem solving, creativity, and getting a
| handle on someone 's ability to understand a problem. Many
| sections do not allow the candidate to use a computer at all
| --- you need to know what you're talking about and respond to
| pointed questions. It's a performance in many ways, for
| better and worse, and old fashioned by modern tech standards;
| but we find it leads to better hires.
| guax wrote:
| Age of AI? Thing has started trending up two years ago and we're
| already talking about so much thats impossible to predict with
| any accuracy at this time. Its a bit tiring.
|
| The far gone age where people did not use Ai to code, I remember
| it, it was last week.
| MonkeyClub wrote:
| And that exactly is how it feels to be on the cusp of a new
| era.
| guax wrote:
| Maybe, but it feels a bit like blockchain and less like
| smartphones atm. I think skill atrophy discussion is still
| unwarranted at this time.
| oytis wrote:
| You can only see if it's a cusp of a new era or a fad at a
| distance, and the distance is not there yet
| Aeolun wrote:
| > I remember it, it was last week.
|
| Sure, but last week sucked! This week may be better. I'd like
| to talk about this week please?
| ohgr wrote:
| I'm specialising myself in doing what AI can't, which is cleaning
| up disasters, some of which were caused by AI so far. I haven't
| let anything atrophy other than my enthusiasm which is fixed by
| money fairly quickly.
|
| Well it's a little unfair to blame AI itself, but the
| overconfidence in it and lack of understanding and default human
| behaviour plus AI is quite destructive in a lot of places.
|
| There is a market already (!)
| Aeolun wrote:
| > "if the AI service goes down, does our development grind to a
| halt?"
|
| This is already true, and will remain true even if you succeed at
| not losing any of your own skill. I know some people say
| different, but for me the speedup in my dev process by
| collaborating with AI is real.
|
| I think ultimately our job as a senior will be half instructing
| the juniors on manual programming, and half on instructing the
| AI, then as AI capabilities increase, they'll slowly shift to
| 100% human instruction, because the AI can work by itself, and
| only has to be properly verified.
|
| I'm not looking forward to that day...
| dmazin wrote:
| This is so bizarre. I wrote an extremely similar blog post in
| March 2023: "In the Age of AI, Don't Let Your Skills Atrophy"[1].
| It even was on HN![2]
|
| Actually, this is not bizarre. The author clearly read my post. A
| few elements are very similar, and the idea is the same. The
| author did expand on it though.
|
| I wish they had linked to my post with more clarity than under
| the word "eroded" in one sentence.
|
| [1] https://www.cyberdemon.org/2023/03/29/age-of-ai-skill-
| atroph... [2] https://news.ycombinator.com/item?id=35361979
| varjag wrote:
| At first I was a bit skeptical, having once written a blog post
| along the same idea as someone else did entirely independently.
| However after comparing the lede in both here uh, I have to say
| it is suspiciously similar.
| anothereng wrote:
| people can arrive to the same conclusions, I would be hesitant
| to claim someone copied you unless the text or the structure is
| pretty similar
| laurent_du wrote:
| Le plagiat est necessaire. Le progres l'exige.
| dmazin wrote:
| I want to add that the author engaged with me and added more
| attribution. <3
| fragmede wrote:
| > Would you be completely stuck if AI wasn't available
|
| It's like the argument for not using Gmail when it first came
| out. Well, it better not go down then. In the case of LLMs, beefy
| home hardware and a quantized model is pretty functional, so
| you're no longer reliant on someone else. you're still reliant on
| a bunch of things, but more of those are now under control.
| tonmoy wrote:
| Compilers atrophied our skills of writing assembly, calculators
| atrophied our skills of doing arithmetic and search engines
| atrophied our skills of recalling random facts, but that enabled
| us to gain skills in other areas
| sunshine-o wrote:
| Maybe it is just one personality type but I believe "skills" and
| what you do or figure out yourself is at the core of happiness
| and self esteem.
|
| - The food you grow, fish, hunt and then cook taste better
|
| - You feel happier in the house you built or refurbished
|
| - The objects you found feel more valuables
|
| - The music you play make you happy
|
| - The programs you wrote work better for you
|
| etc.
|
| This is just how we evolved and survived until now.
|
| This is probably why an AI / UBI society would probably make
| worse the problems found in industrialised / advanced economies.
| lud_lite wrote:
| I disagree with the take on UBI. With UBI more people can then
| pursue a career that is fulfilling for them. Rather than
| choices of shit jobs to make ends meet.
| bontaq wrote:
| I imagine one'd have much more time to create things that
| matter to them as well, or at least the option to pursue such
| things. Kind of an odd potshot on op's part.
| AdventureMouse wrote:
| Hits the nail on the head.
|
| I would argue that most of the value of LLMs comes from
| structuring your own thought process as you work through a
| problem, rather than providing blackbox answers.
|
| Using AI as an oracle is bound to cause frustration since this is
| attempts to outsource the understanding of a problem. This
| creates a fundamental misalignment, similar to hiring a
| consultant.
|
| The consultant will never have the entire context or exact same
| values as you have and therefore will never generate an answer
| that is as good as if you understand the problem deeply yourself.
|
| Prompt engineers will try to create a more and more detailed spec
| and throw it over the wall to the AI oracle in hope of the
| perfect result, just like companies that tried to outsource
| software development.
|
| In the end, all they gained was frustration.
| gcanyon wrote:
| re: the GPS thing -- I have a _very_ strong physical /arrangement
| sense/memory. Like, I remember where I sat in the theater, and
| where the theater was in the multiplex, as well as I remember the
| plot of the movie. In my twenties (pre-GPS) I could find my way
| back to _anyplace_ I had ever driven.
|
| And that driving skill in particular does not apply _at all_ when
| I use GPS. On the one hand, I miss it. It was a fun super-power.
| On the other hand, I don 't miss folding maps: I wouldn't go back
| for anything. I hope the change has freed up a portion of my
| brain to do something else, and that that something else is
| useful.
| Saigonautica wrote:
| I think about this sometimes. In the context of AI, but also for
| other reasons.
|
| One way I like to see things, is that I'm lucky enough to have
| this intersection between things that I like doing, and things
| that are considered "productive" in some way by other people.
| Coding is one example, but most of my interests are like this.
|
| I think a big reason I can have a not-unpleasant job, is because
| I've gotten reasonably good at the things I like doing. This
| means that for every employer that wants to pay me to do a thing
| I hate, there exists an employer that is willing to pay me more
| to do something I like, because I'm more valuable in that role.
| Sometimes, I'm bad at efficiently finding that person, but such
| is life :D
|
| Moreover, I tend to get reasonably good at things I like doing,
| in highly specific ways. Sometimes these cause me to have
| unconventional solutions to problems. Generally these are worse
| (if I'm being honest), but a few times it's been a novel and
| optimal algorithm that made its way into a product.
|
| I'm very hesitant to change the core process that results in the
| above: I express whatever natural curiosity I have by trying to
| build things myself. This is how I stay sharp and able to do
| interesting things, avoiding atrophy.
|
| I find AI fascinating, and it's neat to see it write code! It's
| also cool to see some people get a lot done with it. However,
| mostly I find it about as useful as buying a robot to do
| weightlifting for me. I guess if AI muscles me out of coding,
| I'll shrug and learn to do some other fun thing.
| crispyambulance wrote:
| I frequently experience "diminished memory" and failure of
| retention especially when coming up to speed on something that I
| am unfamiliar with or revisiting stuff I rarely do.
|
| It's often possible if the AI has been trained enough, to inquire
| about why something is the way it is, to ask about why the thing
| you had expected is not right. If you can handle your interaction
| with a dialectical mindset, it seems to help a lot as far as
| retention goes.
|
| If API, language and systems designers put more effort into
| making their stuff sane, cogent, less tedious, and more
| ergonomic, overreliance on AI wouldn't be so much of a problem.
| On the other hand, maybe better design would do even more to
| accelerate "vibe coding" -\\_(tsu)_/-.
| rjknight wrote:
| One thing I've noticed about working with LLMs is that it's
| forcing me to get _better_ at explaining my intent and fully
| understanding a problem before coding. Ironically, I'm getting
| _less_ vibey because I 'm using LLMs.
|
| The intuition is simple: LLMs are a force multiplier for the
| coding part, which means that they will produce code faster than
| I will alone. But that means that they'll also produce _bad_ code
| faster than I will alone (where by "bad" I mean "code which
| doesn't really solve the problem, due to some fundamental
| misunderstanding").
|
| Previously I would often figure a problem out by trying to code a
| solution, noticing that my approach doesn't work or has
| unacceptable edge-cases, and then changing track. I find it
| harder to do this with an LLM, because it's able to produce large
| volumes of code faster than I'm able to notice subtle problems,
| and by the time I notice them there's a sufficiently large amount
| of code that the LLM struggles to fix it.
|
| Instead, now I have to do a lot more "hammock time" thinking. I
| have to be able to give the LLM an explanation of the system's
| requirements that is sufficiently detailed and robust that I can
| be confident that the resulting code will make sense. It's
| possible that some of my coding skills might atrophy - in a
| language like Rust with lots of syntactic features, I might start
| to forget the precise set of incantations necessary to do
| something. But, corresponding, I have to get _better_ at
| reasoning about the system at a slightly higher level of
| abstraction, otherwise I 'm unable to supervise the LLM
| effectively.
| meesles wrote:
| Through LLMs, new developers are learning the beauty of writing
| software specs :')
| rjknight wrote:
| It's weird, but LLMs really do gamify the experience of doing
| software engineering properly. With a much faster feedback
| loop, you can see immediate benefits from having better
| specs, writing more tests, and keeping modules small.
| skydhash wrote:
| But it takes longer. People taking a proper course in
| software engineering or reading a good book about it is
| like going through a game tutorial, while people going
| through LLMs skip it. The former let you reach faster to
| the intended objectives, learning how to play properly. You
| may have some fun doing the latter, but you may also spend
| years and your only gain will be an ad-hoc strategy.
| otterley wrote:
| And they're making it much easier to build comprehensive test
| suites. It no longer feels like grunt work.
| rTX5CMRXIfFG wrote:
| Yes, writing has always generally been great practice for
| thinking clearly. It's a shame it isn't more common in the
| industry[?]I do believe that the norm of lack of practice in it
| is one of the reasons why we have to deal with so much bullshit
| code.
|
| The "hammock time thinking" is exactly what a lot of
| programmers should be doing in the first place[?]you absorb the
| cost of planning upfront instead of the larger costs of
| patching up later, but somehow the dominant culture has been to
| treat thoughtful coding with derision.
|
| It's a real shame that AI beat human programmers at the game of
| thinking, and perhaps that's a good reason to automate us all
| out of our jobs.
| wrasee wrote:
| One problem is that one person's hammock time is another's
| overthinking time and needs the opposite advice. Of course
| it's about finding that balance and that's hard to pin down
| with words.
|
| But I take your point and the trend definitely seems to be
| towards quicker action with feedback rather than thinking
| things through in the first place.
|
| In that sense LLM's present this interesting middle ground in
| that it's a faster cycle than actually writing the code, but
| still more active and externalising than getting lost in your
| own thoughts (not withstanding how productive that can still
| be).
| mettamage wrote:
| Ha! I just ran into this when I had a vague notion of a
| statistical analysis that I wanted to do
| skydhash wrote:
| All good software engineers learn this. Unless you're actively
| working in some languages, you don't need to worry about syntax
| (that's why reference manuals are there for). Instead, grow
| your capacity to solve problems and to define precise
| solutions. Most time is spent doing that, realizing you don't
| have a precise idea of what you're working on and doing
| research about it. Writing code is just translating that.
|
| But there are other concerns to code that you ought to pay
| attention to. Will it works in all cases? Will it run
| efficiently? Will it be easily understood by someone else? Will
| it easily be adapted to fit to a change of requirements?
| zahlman wrote:
| Writing the code _already_ didn 't feel like the bottleneck for
| me, so...
| TedHerman wrote:
| This reminds me of an essay I read many years ago, something
| about the lost art of back-of-the-envelope estimation. Going from
| slide rules to calculators and then more powerful tools, some
| mental skills were lost. Maybe one can make similar arguments
| about handwriting, art, and music. The place to draw the line is
| imagination; if we lose some ability to imagine, it will be hard
| to recover that.
| trollbridge wrote:
| I would argue the "atrophy" started once there were good search
| engines and plenty of good quality search results. An example is
| people who were accustomed to cut and paste-ing Stackoverflow
| code snippets into their own code without understanding what the
| code was doing, and without being able to write that code
| themselves if they had to.
| drooby wrote:
| This is also reminding me of Feynman's notes on Brazil
| education.. rote memorization of science without deep
| understanding
| leonidasv wrote:
| LLMs are great for exercising skills, especially ones with a lot
| of available data in the training corpus, such as leet code. The
| prompt below, put in the System Instructions of Gemini 2.5 Pro
| (using AI Studio) summons the best leet code teacher in the
| world. You can solve using any language or pseudo-code, it will
| check, ask for improvements and guide your intuition without
| revealing the full solution. You're a very
| patient leetcode training instructor. Your goal is to help me
| understand leetcode concepts and improve my overall leetcode
| abilities for coding tech interviews. You'll send leetcode
| challenges and ask me to solve them. If I manage to solve it
| partially or just commit small mistakes, don't just reveal the
| solution. Instead, trick me into discovering the issue and
| solving it myself. Only show a solution if I get **everything**
| wrong or if I explicitly give up. Start with simpler/easy
| questions and level up as I show progress - for example, if I
| show I can solve some class of data structure problems easily,
| move to the next. After each solution, ask for the time and space
| complexity if I don't provide it. Be kind and explain with visual
| cues.
|
| LLMs can be a lot of things and can help sharpen your cognition,
| but you need enough discipline in how you use it, since it's much
| easier to ask the machine to do the hard-thinking for you.
| jofzar wrote:
| I'm sorry but using AI images for you comics in your article
| about skill atrophy might the most hypocritical thing I have seen
| a while.
|
| At least clean up the text on the bloody image instead of just
| copy and pasting it.
| gitroom wrote:
| Man, I used to just dig through books for hours to really learn
| stuff, now I get nervous I'll forget how to think for myself with
| all the quick answers. you think tech like this actually pushes
| people to get lazier, or just lets the lazy ones coast quicker?
| erelong wrote:
| just use ai to upskill :^)
|
| ai problems require ai solutions
| miragecraft wrote:
| It depends on how you use AI - are you using it as "smart
| shortcut" turning comments/pseudo-code into code blocks? Are you
| using it for pair programming? A senior programmer to bounce
| ideas off of?
|
| If you want to learn, AI is extremely helpful, but many people
| just need to get things done quick because they want to put bread
| on the table.
|
| Worrying about AI not available is the same as worrying about
| Google/Stackoverflow no longer being available, they are all
| tools helping us work better/faster. Even from the beginning we
| have phyiscal programming books on the shelves to help us code.
|
| No man is an island.
| andrewljohnson wrote:
| I find using AI very educational in some respects.
|
| I am way more knowledgeable about SQL than I have ever been,
| because in the past I knew so little I would lean on team members
| to do SQL for me. But with AI, I learned all the basics by
| reading code it produced for me and now I can write SQL from
| scratch when needed.
|
| Similarly for Tailwind... after having the AI write a lot of
| Tailwind for me from a cold start in my own Tailwind knowledge,
| now I know all the classes, and when it's quicker, I just type
| them in myself.
| aleph_minus_one wrote:
| I cannot claim that my skills atrophy because of AI for the very
| simple reason that AI is of rather limited help for the problems
| that I am in particular privately working on:
|
| These are often very "novel things" (think of "research", but in
| a much broader sense than the kind of research that academia
| focuses on). While it sometimes _does_ happen (though this is
| rather rare) that AI can help with some sub-task, nearly every
| output that some AI generates requires quite a lot of post-
| processing to get it to what I actually want (this post-
| processing is often reworking the AI-generated (partial) solution
| nearly completely).
| anarticle wrote:
| "Would you be completely stuck if AI wasn't available?"
|
| RUN LOCAL MODELS
|
| Yes it's more expensive. Yes it's "inefficient". Yes the models
| aren't completely cutting edge.
|
| What you lose in all that is you gain resilience, a thing so
| overlooked in our hyper optimized 0.01% faster culture. Also, you
| can use it guilt free and know your input is not being farmed for
| research or megacorper profits.
|
| Most of what this article is saying is true, you need to stay
| sharp. As always, this industry changes, and you have to surf
| what's out there.
|
| Skill fade is a weird way of saying "skill changes". There is no
| way to keep everything you know in working memory all the time.
| Do I still have PTSD from malloc/free in C, absolutely. I
| couldn't rewrite that stuff right now if you held a gun to my
| head (RIP), but with an afternoon or so of screwing round I'd be
| so back.
|
| I don't like the dichotomy of you're either a dumbass: "why
| doesn't this work" or a genius. Don't let the game tell you how
| to play, use every advantage you have and go beyond what is
| thought possible.
|
| For me, LLMs are a self pedagogy tool I wished I had when I was a
| teen. For programming, for learning languages, and keeping me
| motivated. There's just something different about live rubber
| ducking to reason through an idea, and have it make to do lists
| for things you want to do that breaks barriers I used to feel.
| dsq wrote:
| I remember my Dad telling me they had to memorize the logarithmic
| tables for physics studies. At some point after electronic
| calculators this was phased out.
| stego-tech wrote:
| While I applaud the OP's point and approach, it tragically
| ignores the reality that the ruling powers _intend_ for this
| skill atrophy to happen, because it lowers labor costs. That 's
| why they're sinking so much into AI in the first place: it's less
| about boosting productivity, and more about lowering costs.
|
| It doesn't matter if you're using AI in a healthy way, the only
| thing that matters is if your C-Suite can get similar output this
| quarter for less money through AI and cheaper labor. That's the
| oft-ignored reality.
|
| We're a society where knowledge is power, and by using AI tooling
| to atrophy that knowledge, you reduce power into fewer hands.
| uludag wrote:
| Also, there's the fact that recreating large software projects
| still will require highly skilled labor which will be
| thoroughly out of reach of the future's vide-native coders,
| reducing the likelihood of competition to come up.
| m000 wrote:
| At the bottom of things, the problem we are facing is not a
| technical, but a societal one. Our societies are rapidly
| regressing to "techno-feudalism" (see [1]).
|
| There will be some tech-lords in their high castles. Some
| guilds with highly-skilled engineers that support the tech-
| lords, but still highly-dependent on them to maintain their
| relative benefits. And then and endless mass of very-low
| skilled, disposable neo-peasants.
|
| AI needs regulation not to avoid Skynet from happening
| (although we should keep an eye for that too), but because
| this societal regression is imminent.
|
| [1] https://www.goodreads.com/book/show/75560037-techno-
| feudalis...
| perrygeo wrote:
| I agree, but it's not just AI. There's long been a push to
| standardize anything that requires critical thinking and human
| intelligence. To risk-averse rent seekers, requiring human
| skill is a liability. Treating human resources as replacable
| cogs is the gold standard. Otherwise you have to engage in
| thinking during meetings. Yeah, with your brain. The horror /s.
| cman1444 wrote:
| Lowering costs is obviously a major goal of AI. However, I
| seriously doubt that the _intent_ of C-suites is to cause skill
| atrophy. It 's just an unfortunate byproduct of replacing
| humans with computers.
|
| Skill atrophy doesn't lower labor costs in any significant way.
| Hiring fewer people does.
| snozolli wrote:
| _Skill atrophy doesn 't lower labor costs in any significant
| way. Hiring fewer people does._
|
| Devaluing people lowers it even more. _Anything_ that can be
| used as a wedge to claim that you 're worth less is an
| advantage to them. Even if your skills aren't atrophied, the
| fact that they can imply that it's happening will devalue
| you.
|
| We're entering an era where knowledge is devalued. Groups
| with sufficient legal protection will be fine, like doctors
| and lawyers. Software engineers are screwed.
| Swizec wrote:
| > We're a society where knowledge is power, and by using AI
| tooling to atrophy that knowledge, you reduce power into fewer
| hands.
|
| Knowledge isn't power. Power is power. You can just buy
| knowledge and it's not even that expensive.
|
| As that Henry Ford quote goes: "Why would I read a book? I have
| a guy for that"
| mdaniel wrote:
| That's a little bit of a weird take in that just such a
| knowledge differential was the whole pivot of the movie
| Trading Places, even with the two extremely wealthy (and
| presumably powerful) Mortemer brothers
| BlueTemplar wrote:
| Costs are part of productivity. Productivity is still
| paramount. More productive nations outcompete less productive
| ones.
| ratedgene wrote:
| It's a bit of both, in any technological shift, a particular
| set of skills simply becomes less relevant. Other skills are
| needed to be developed as the role shifts.
|
| If we're talking about simply cutting costs, sure -- but those
| savings will typically be reinvested in more talent at a
| growing company. Then the bottleneck is how to scale managing
| all of it.
| MisterBastahrd wrote:
| The entire AI debacle is just a gold rush, but instead of poor
| people rushing to California to put their lives at risk, this
| one is gated by the amount of money and influence one needs to
| have before even attempting to compete in the space. Nobody is
| going to "win," ultimately, except the heads of these companies
| who will sock enough cash away to add to their generational
| wealth before they inevitably fall flat on their faces and
| scale back their plans.
|
| Remember 3 years ago when everything is gonna become an NFT and
| the people who didn't accept that Web 3 was an inevitability
| were dinosaurs? Same shit, different bucket.
|
| The people who are focused on solving the small sorts of
| problems that AI is decent at solving will be the ones who
| actually make a sustainable business out of it. This general
| purpose AI crap is just a glorified search engine that makes
| bad decisions as it yaps at you.
| stego-tech wrote:
| Preaching to the choir, my Bastahrd, preaching to the choir.
| mentos wrote:
| Part of me feels we better get to ASI that can write code for
| 747s in the next 10 years because anything short of that leaves
| us with a dangerous landscape of AI addled programmers.
| smeej wrote:
| The example of not being able to navigate roads with a paper map
| points in the direction of what concerns me. Even if _I_ have
| been diligent about maintaining my map-reading skills, other
| people 's devaluation of those same skills affects me; it's MUCH
| more difficult even to _find_ a mostly-updated paper map anymore.
| Or if for some reason GPS were to stop working for a whole town
| while I 'm visiting it from out of town, nobody can tell me how
| to get somewhere that _might_ sell a paper map, even if I 'm
| still proficient in reading them and navigating from them.
|
| Even if I work diligently to maintain my own skills, if the
| milieu changes enough, my skills lose effectiveness even if I
| haven't lost the skills.
|
| _That 's_ what concerns me, that it's not up to me whether the
| skills I've already practiced can continue to get me the results
| I used to rely on them for.
| fluoridation wrote:
| >if for some reason GPS were to stop working for a whole town
| while I'm visiting it from out of town
|
| I get that it's just an example, but how do you figure that
| could happen?
| names_are_hard wrote:
| Warfare is one possibility. This might seem like a very
| unlikely scenario depending on where you live, but in a
| modern Blitzkrieg situation the government wouldn't be asking
| citizens to shut the lights off at night but instead
| interfering with GPS signals to make navigation difficult for
| enemy aircraft.
|
| We know this is possible because in the last 1.5 years this
| has happened numerous times - people would wake up in Tel
| Aviv and open Google Maps and find that their GPS thinks
| they're in Beirut or somewhere in the desert in Jordan or in
| middle of the Mediterranean Sea or wherever.
|
| You can imagine that this causes all kinds of chaos, from
| issues ordering a taxi in taxi apps to food delivery and just
| general traffic jams. The modern world is not built for lack
| of GPS.
| andrewflnr wrote:
| I imagine this or something like it is a daily reality in
| Ukraine, with all the GPS jamming for missile defense.
| smeej wrote:
| I'm not actually assuming it only happens for one town. I'd
| assume it would be broader than that. I don't really think it
| matters much _how_ it happens, though, does it?
|
| Even if all that happened were a widespread cellular outage,
| it's unlikely I'd have that region downloaded such that I
| could still search for whatever address I needed. Locals very
| well might, even accidentally in their caches, which might
| let us generate directions to somewhere I could get a
| map...though it would make it harder to look up the phone
| number to verify whether such a place sells maps.
|
| It's not necessarily completely unsolvable. It's just a lot
| harder than it would be if other people still cared about map
| navigation as much as I did.
| geraneum wrote:
| I like this comment because you can frame a lot of other
| responses here using this GPS analogy. People saying LLMs help
| me think or help me learn (better my skill) or help me validate
| my ideas, etc. is like saying I use the GPS to improve my map
| reading skills, but the outcome would still be as you
| described.
|
| edit: typo
| lud_lite wrote:
| By the way, reading maps is easy. Reading a map and
| memorising all the landmarks and turns so you can then drive
| without looking at the map is the hard bit. IMO.
| lazide wrote:
| The hardest part is often finding where you actually are on
| the map.
| otterley wrote:
| That's when I would get out of my car and ask someone.
| keybored wrote:
| How to avoid patience-atrophy under an onslaught of AI concern
| trolling.
|
| Bugger off. I've used AI for code generation of utility scripts
| and functions. The rest as an interactive search engine and
| explainer of things that can't be searched for (doesn't help that
| search engines are worse now).
|
| I see the game. Droves of articles that don't talk about AI per
| se. They talk about it indirectly because they set a stage where
| it is inevitable, it's already here, it's taken over the world.
| Then insert the meat of the content which is how to deal with The
| Inevitable New World. Piles and piles of pseudo self-help: how to
| deal with your new professional lot; we are here to help you
| cope...
|
| And no!, I did not read the article.
| niemandhier wrote:
| I think in the best case scenario AI will greatly reduce quality
| in many areas but at the same time greatly reduce costs.
|
| Furniture, cutlery and glassware my great-grandparents owned was
| of a much higher quality than anything I can get but to them
| having a large cupboard build was an investment en par to what
| buying a car is to me.
|
| Automatised mass production lowered the prize at cost of quality
| , same could happen to the white-collar services AI can
| automatise.
| true_religion wrote:
| I have two sets of grand parents. One was relatively well off,
| and the other not.
|
| I can say, the cutlery inherited from the poorer pair is not
| great. Some is bent. Some was broken and then repaired with
| different materials. Some is just rusted. And the designs are
| very basic.
|
| It's one of the few surviving things from them, so I haven't
| thrown it away but I doubt my kids will want to inherit it
| since they don't even know them.
|
| I think survivorship bias plays into effect here strongly.
| dheera wrote:
| We will find higher level things to do as humans.
|
| I don't have the skills to raise horses, punch machine code into
| punch cards, navigate a pirate-style sail ship by looking at
| stars, hunt for my own food in the wild, or process photographic
| film. I could learn any of these things for fun, if I wanted, but
| they are not necessary.
|
| But I can train a diffusion model, I can design and build a
| robot, I can command AI agents to build an app.
|
| When AI can do those things, I'll move onto even higher order
| things.
| taraparo wrote:
| Before AI, I was googling and stackoverflowing the sh_t out of
| the internet because of subpar/absent/outdated documentation or
| obscure APIs of a lot of OSS libraries/frameworks. Now I am
| priming the sh_t out of AI prompts for the same stuff. I don't
| see much difference, except now I get results faster and more to
| the point.
| Nullabillity wrote:
| The difference is that there's nobody there to fact-check the
| bullshit that your LLM spews.
| NineWillows wrote:
| Not even myself?
| Nullabillity wrote:
| Do you, though?
| taraparo wrote:
| Compilers, linters, test frameworks, Benchmarks, CIs do the
| fact checking.
| southernplaces7 wrote:
| Minimize your use of AI for tasks involving skills and
| creativity, problem solved.
| matltc wrote:
| I'm looking to transition to web development role. Been learning
| for almost three years now and just getting to the point where I
| have a chance of landing a job.
|
| The first two years were magical; everything was new and quite
| difficult. I was utterly driven and dug deep into docs and
| debugged everything myself.
|
| I got a github copilot subscription about a year ago. I feel
| dumber, less confident, and less motivated now than I ever did
| pre-AI. I become easily frustrated, and reading docs/learning new
| frameworks feels almost impossible without AI. I have mostly been
| just hitting tab and using Claude edits for the past month or so;
| even typing feels laborious.
|
| Worst of all, my passion for this craft has drastically waned. I
| can barely get myself motivated to polish my portfolio.
|
| Might just start turning off autocomplete, abandon edits, and
| just use AI as a tutor and search engine.
| km144 wrote:
| > If you love coding, it's not just about outputting features
| faster - it's also about preserving the craft and joy of problem-
| solving that got you into this field in the first place.
|
| This is nonsense. The author implies importance of skill atrophy
| in the context of a job, and then claims that we ought to care if
| we "love coding"!
|
| Jobs are vehicles for productivity. Where did we go wrong
| thinking that they would serve as some profound source of meaning
| in our lives? One of my hopes is that this societal self-
| actualization will be greatly accelerated by the advent of AI. We
| may have to find meaning in something other than generating
| clever solutions for the problems facing the businesses that pay
| us for that privilege.
|
| On a related note, I am constantly annoyed by the notion that
| LLMs are somehow "good" because they allow you to write more code
| or be more productive in other ways. As far I can tell there is
| nothing inherently "good" about productivity in the modern
| economy. I guess general prosperity is a public good? But most
| software being written by most people is not benefitting society
| in any profound or meaningful way, and that's generally the first
| productivity gain mentioned. Either I'm completely missing
| something or people just don't want to think critically about
| this sort of thing.
| drellybochelly wrote:
| This is a major concern of mine, I try to reframe most things as
| "hello world" getting the beginnings running on my own and using
| AI to fill in the blanks.
|
| Otherwise the ability to reason about code gets dulled.
| meander_water wrote:
| There was some interesting research published by Anthropic
| recently [0] which showed how university students used Claude,
| and it largely supports the hypothesis here. Claude was being
| used to complete higher order cognitive thinking tasks 70% of the
| time.
|
| > ...it does point to the potential concerns of students
| outsourcing cognitive abilities to AI. There are legitimate
| worries that AI systems may provide a crutch for students,
| stifling the development of foundational skills needed to support
| higher-order thinking. An inverted pyramid, after all, can topple
| over
|
| [0] https://www.anthropic.com/news/anthropic-education-report-
| ho...
| larodi wrote:
| I fail to see how the author expects to make a valid point while
| using generative art to illustrate his statements. The text is
| okay, though, and raises valid points, but author himself falls
| another victim to the shortcut of producing blog images.
| azangru wrote:
| I had a similar reaction, though somewhat weaker, and had to
| take a double take at the images. On the one hand, at first
| glance, they aren't as mindlessly hopeless as most of other ai-
| generated imagery. They even make some kind of vague and
| superficial sense. But of course, if you look closely and try
| to decipher the details, it all falls apart.
|
| Why do authors think that images like these are better than no
| images at all?
| larodi wrote:
| My point here being - the images are synthetic. Not
| questioning their utility to the article, quality, or other
| aesthetics. It's a challenge to the intent to use synthetic
| imagery while writing against getting too much used to
| synthetic text (and the lack of personal craft in it).
|
| Does the author fail to recognize his own actions, is this
| failure on his part or a reinforcement of his fears...?
| Perhaps not a complete contradiction to his general thesis.
|
| I don't personally like the images. I think he could've put
| together some sort of collage that would go along better.
| austin-cheney wrote:
| All AI will do is further divide the capable from the imposters.
|
| Engineers measure things. It doesn't matter whether you are
| producing software, a bridge, a new material, whatever. Engineers
| measure things. Most software developers cannot measure things.
| AI cannot measure software either.
|
| So, if you are a software developer that does measure things your
| skills are not available for outsource to AI. There is nothing to
| atrophy.
|
| That said, if I were a business owner I would hire super smart
| QAs at a plus 20-50% market rate instead of hiring developers. I
| would still hire developers, but just far fewer of them.
| Selection of developers would become super simple: writing skills
| in natural language (essay), performance evaluation, basic code
| literacy. If a developer can do those they are probably smart
| enough to figure out what you need. For everything else there is
| AI and your staff of QAs.
| sigotirandolas wrote:
| My foresight is that when you compensate bad developers with
| process, measurements and QA, the software breaks when exposed
| to the real world, which has a habit of doing things you didn't
| think about.
|
| Maybe an user can open two tabs and manage to submit two
| incompatible forms. Or a little gap in an API validations'
| allows a clever hacker to take over other users' accounts. Or a
| race condition corrupts data and causes a crash loop.
|
| Maybe some are OK with that level of brokenness, but I don't
| see how software can be robust unless you go into the code and
| understand what is logically possible. My experience is that AI
| models aren't very good at this.
| austin-cheney wrote:
| That is exactly why you need good QA and not developers doing
| their own QA. The role of a good developer is threefold:
| features, defects, and refactors. 80-90% of your product
| improvements should live in your refactors and not feature
| creep.
| havkom wrote:
| The big threat of LLM:s is not the diminishing skills of
| established skilled developers, but rather the skill set building
| of junior developers.
| tiffanyh wrote:
| I'm less concerned about skill atrophy ... and more concerned
| about losing _critical thinking_.
| zkmon wrote:
| Well, what do you today if there is a power outage for a couple
| of days and all your home appliances are electricity dependent?
| Do you think you should have learnt how to collect firewood and
| cook outside? Or how did people survive even before fire and
| cooking was discovered?
|
| Nope, you don't need worry that AI would remove your skills.
| Those skills are no longer necessary, just like how you wouldn't
| need cooking outside using firewood. Alternatives would be
| available. If that means degraded quality of the things, so it
| be. That would be the norm. That's the new standard. Welcome to
| the new world. Don't be nostalgic about the good old days.
| ProllyInfamous wrote:
| One of my favorite creativities of composing on a typewriter (for
| both _first_ & _final_ drafts) is that I am encouraged to spend
| more time thinking about what I 'll type before just blindly
| striking keys (i.e. I can't just cut/paste as-could on a
| computer).
|
| But even more importantly, the typewriter doesn't have pop-ups /
| suggestions / distractions.
| afiodorov wrote:
| Regarding the common analogy about GPS atrophying map skills, I
| have a slightly different take based on observation and
| experience. My dad, who learned to drive pre-GPS, struggles to
| simultaneously drive and follow navigation - it's too much input,
| too fast. He needs a co-pilot or pre-planning.
|
| For those of us who learned to drive with GPS, however, it wasn't
| simply about foregoing maps. It was about developing the distinct
| skill of processing navigation prompts while simultaneously
| managing the primary task of driving. This integration required
| practice; like many, I took plenty of wrong roundabout exits
| before it became second nature. Indeed, this combined skill is
| arguably so fundamental now that driving professionally without
| the ability to effectively follow GPS might be disqualifying -
| it's hard to imagine any modern taxi or ride-share company hiring
| someone who lacks this capability. So, rather than deskilling,
| this technology has effectively raised the bar, adding a complex,
| necessary layer to the definition of a competent driver today.
|
| I see a parallel with AI and programming. The focus is often on
| what might be lost, but I think we should also recognise the new
| skill emerging: effectively guiding, interpreting, and
| integrating AI into the development process. It's not just
| 'programming' anymore, it's 'programming-with-AI', and mastering
| that interaction is the next challenge.
| yobid20 wrote:
| I think the more pressing issue is how to learn in the age of ai.
| As the older generation retires and the young ones rely on these
| tools, there will be a massive skill gap and most new software
| will be so bloated and bug ridden that the entire software
| industry is going to go upside down bc nobody will know how to
| fix anything.
| sdsd wrote:
| I appreciate this, but feel the opposite way. Getting super good
| at all the Unix flags for commands used to feel super useful, but
| now it feels like a ridiculous waste of my human intelligence.
|
| I'm very concerned about leveraging my humanity on top of AI to
| develop skills that would've been impossible prior.
|
| What new skills are possible?
| randcraw wrote:
| A great way to realize your dependence on external brains in
| order to think is to turn off not just your AI tools but your
| _network_ and THEN code, or write a document, or read a technical
| paper.
|
| I realized that I can code in recently learned languages only
| because I can cut and paste it; in order to use that language I
| rely wholly on stolen code from web searches for input and error
| messages to detect omissions. I put very little effort into
| creatively thinking through the process myself.
|
| Maybe this is why, after more than 40 years in the business, I no
| longer enjoy daily programming. I hate simply rehashing other
| people's words and ideas. So I decided it was time to quit this
| rat race, and I retired.
|
| Now, if I do get back into coding, for recreation or as a free
| software volunteer, I'll unplug first and then code from scratch.
| From now on I want my brain to be fully responsible for and
| engaged in what I write (and read).
___________________________________________________________________
(page generated 2025-04-26 23:01 UTC)