"People who don't use AI will be left behind"
migrainebrain.bearblog.dev
migrainebrain.bearblog.dev
I'm very thankful I came of age during the golden age of personal computing. I was able to own my own computer(s) and earn a living writing software on them and for them. Fifty years was a good run, and I consider myself lucky to have participated in it.
IMO we've gone full circle: dumb terminals chained to mainframes and the whimsey of someone else's rules, restrictions, and rent-seeking, to my own bought-and-paid-for computer sitting on my desk that did exactly what I told it to do using software that never changed unless I wanted it to change, and now we're back to dumb terminals (browsers) that talk to mainframes (the cloud) that not only harvest and sell my personal information to the highest bidder but constantly change the rules and restrictions on my software and have gone back to renting me the software and pushing changes that I never asked for and never wanted in the first place.
I will never use spicy autocomplete for anything, and I find it depressing that people are being forced to use it in order to keep their job. I see a very dark future for computing if real skills are all replaced with garbage being vomited out by rules engines that harvested their "guess the next word" results from today's internet.
In the long arc of history, I'm doubtful we'll see 'the last 50 years' as 'the Golden Age' - that's just a personal, contemporary romanticization. More than likely, the advent of computers -> web -> AI etc. will be one block of the 'informational industrial revolution'.
The people who made the ostensible 'Golden Era' were pioneers, just as those breaking new ground are pioneers today, it's honestly 'depressing' that people who consider themselves 'Engineers' wouldn't see that as clear as day, an be hopeful for the future on some level.
AI is very real phenom, obviously vastly over-hyped in many ways, and it doesn't feel nice to have to get caught up in a tectonic shift against one's will, but it is bringing about legitimate progress in every sense that the Engineers and Creators before us did.
In the exact same spirit as DaVinci or Babbage.
If one wants to keep a horse in the stable, or a typewriter around for posterity or any other reason that's fine, but not under the notion that somehow they are better or more useful.
Nothing the parent said was arrogant.
Also, if you want to 'go there' you could find a much better word than 'immature' to say what you're trying to say.
The OPs posture is not tonally arrogant, but it's it's definitely intellectually arrogate.
The OP claiming heritage of the 'Golden Era' which is a dramatic, egoic romanticization.
To place one's 'own story at primacy' above all others, insinuating that 'his skills' are those which are 'true and relevant' and that those using new tools are 'lesser' or 'not substantial' , and also grossly myopic to the truly great Engineering that's going on ... is arrogant and insulting frankly.
We can empathize with having to yield to a changing world, or being too out of scope to even fathom the 'new tech', but that's very different with saying that 'Frank Sinatra was the Only Great Singer, those that came after him had not talent'.
If it were the case, then fine, but it's obviously not. AI is a legitimate advancement that narrow minded people are struggling to fathom, and it's coming out in some ugly ways.
A true creator would probably take magnanimous position that after having made their contribution, they are sad to not be able to participate in what is maybe an even more substantial era of progress, and all of the wonders that will come of it.
Good gripes - we're all about to have robots in our homes (!) probably within 5-15, we're witnesses Sci Fi unfold in front of us ...
I do hope the open weight models keep distilling the frontier models, and that powerful and unlocked computer hardware remains accessible to us mere mortals so that we may run them with no limitations in our own homes. That's optimistic though.
Thanks to Chinese vendors, there's a ton of near Tier1 AI that we can all access.
Unlike Android where it's not feasible to use though it may be Open on some level.
They're capitalist entities trying to capture value and control of a new thing they are introducing - yes - but that's a pretty normal thing, we get over that with competition, regulation etc.
This will happen over time - all device manufacturers, cloud providers are 'on our side!'
AWS and Apple want to commoditize the AI as much as you do.
So do the Chinese manufacturers, and really Nvidia and ASML as well.
So we, the proletariat can 'pit them against each other' in value chain competition!
The fact of the matter is, a person working with a bunch of agents is a lot more productive than just a person. It makes research faster. It makes experimentation faster. It makes output cleaner. And this is true across many disciplines, not just tech.
Also, it is a skill. Yes, anyone can chat with an LLM. But understanding the optimal work flow for what to delegate and what to do yourself is difficult. Understanding the need for precision in the language used, and learning how to elegantly phrase things that were previously just abstract thoughts is absolutely a talent that can be refined.
If i had to guess, I'd say we'll probably see major breakthroughs across multiple disciplines within the next decade, largely because researchers and engineers can cover much more ground individually now, freed from the slow moving coordination mechanisms that team dynamics require. Pretty good for "spicy autocomplete" as you put it.
There are jobs outside of IT. They are harder, they have less benefits, they pay less. It's a whole project to switch your lifestyle so you can even afford them.
I know nobody who regrets making the jump. I hope to make it within this year. I'll be poor, but at least I won't work in IT.
> But understanding the optimal work flow for what to delegate and what to do yourself is difficult.
No it's not, you can learn it in less than a day. I've done it a few times while evaluating how much the agents have progressed (despite what people keep saying, not much).
> Understanding the need for precision in the language used, and learning how to elegantly phrase things that were previously just abstract thoughts is absolutely a talent that can be refined.
Some of us learned technical writing to communicate with _humans_ before, and we're sitting here alternating crying and laughing as y'all scramble to figure it out just to put all that into a hallucination machine.
What's your plan?
It's gonna be very broke, but I'm not the first one in my friends circle to make the jump, so I have some support.
Edit: I probably will keep coding. Just... nobody else is ever going to see or use my code again.
If your worry is that you won’t be able to “keep up” and you’ll be laid off, or fired, just wait for that to happen. Keep making a paycheck until then. Then you can start your barista job.
If the problem is that you hate the work, then fine. But why barista? Fine, if that’s what makes you happy. But there are a million jobs out there _if you are willing to relocate_.
Also I'm sick and tired of working on projects where the best social benefit from my work would be if I stopped. And IT has this talent of doing this to even most superficially useful projects. I worked on solar panel software that got turned into a scam by marketing. That takes a talent, of sort.
The best time to jump out of IT was to never get into it. The second best time is now.
As for why barista? People need food and drink and coffee is great.
It depends on where you land. Not all programmers (and their managers) are brain-amputated zombies. But I do admit that finding that rare pocket of sanity requires a good portion of luck.
And if I ever find money under the mattress, I’ll make a solar farm. Something useful for the world, for once.
Better content and poor than living in golden handcuffs.
We have yet to see hard numbers on time saved by those who use LLM tooling extensively. It could be it doesn't turn out as compelling as we might expect.
Just sayin', I never forced software developers to use NetBeans or Intellij IDEA. I'm certainly not changing my tune and forcing them to use LLM tooling either.
Dumb example: graphical user interfaces. Heavyweight IDEs used to have a GUI designer (Netbeans had a very nice one).
GUI development is niche nowadays.
Also we have much better cross-editor tooling, just think of language servers (https://microsoft.github.io/language-server-protocol/) and build servers (https://build-server-protocol.github.io/). Back in the day each IDE had their own.
Vim and Emacs can do a lot of what IDEs used to offer thanks to language servers and build servers. Before those (lang/build servers) they were largely useless for large scale development (believe me, i tried).
Especially for visualising data, to just get a quick look at, it can now be done in lightning speed and I am quite familiar with "manual data processing" in a few languages.
AI use me decently outcompetes manual me. Sometimes there are also stupid tasks. Data must be serialised in a certain way for some stupid reason. You prepare the info to be digested and the busywork can be done by agents with little oversight, which otherwise would have taken you a few hours. There simply is a limit in how fast you can read, look up field names, etc... If you outsource these critical paths to AI, you can gain productivity.
I also start way more side projects now and I like manual coding a lot.
Judging a programmer by how much code they write is like judging an airplane engineer by how heavy their planes are.
That said, if you get something quite heavy to lift off, it probably is a decent engineering feet. Not saying it would be the best plane :)
[citation needed]
I try LLMs for something every couple of months, and I have yet to see them produce anything actually correct. Calling non-existing library methods, confabulations, etc.
But sure, they produce a lot of stuff in a short while. The utility of any of that another question.
That's too pessimistic, the productivity gains are real and substantial.
OTOH, the hype train is out of control. It is nowhere near perfect and requires a lot of handholding and guardrails to avoid going sideways.
You need to adopt it to stay relevant, but don't fall for the excessive hype. At the end of the day the limitations are significant.
Source, please.
Even if its self reported, its useful to know.
As for my source, it is my experience at work for both myself, and my direct reports, and my peer teams.
Tiny example: a certain recurring task we need to do to help other teams which requires somewhat tedious analysis but rarely a lot of high level thinking, plus a bit of decision making. A year ago I'd do maybe 2-3 per day because each one took about 30 minutes so I had to find slices of uninterrupted time between other work to complete them.
Just tonight as I was wrapping up the day I did 19 of these in an hour while also catching up on email. I let the bots do all the research in parallel, as each one completed the research I'd either tell it "ok, do it", or if it was asking me for a decision I'd decide and tell it how to proceed.
I count myself as an AI skeptic, in the sense that the hype is way above reality. But that doesn't mean there isn't a huge amount of real gains.
Douglas Adams, The Salmon of Doubt
"I strongly feel that this is an insult to life itself." -Hayao Miyazaki
On the other hand, the HHGG universe is just packed to the brim with deranged robots. Everybody loves Marvin, of course, but my favorites were the sycophantic ones like the elevators that sigh with pleasure upon delivering you to your destination. Adams always seemed to do perfectly anticipate the insanity of marketeers, and I expect that we’ll actually get some of this someday…
Not sure why LLMs feel the opposite. Maybe it's because of the terrible marketing and pushing it down everyone's throats. Maybe it's because of the personality of people like sama, or how it's being used to produce the so-called AI slop globally. Maybe something completely different. But there's something bleak and off-putting in it.
But it has been so frustrating to listen to the hype and then contend with the actual results.
How bizarre is this tool, that I have been forced to conduct interviews across all domains, to see where people are having actual productivity gains?
I never had to do this with the iPhone, or with word, excel or any number of cool technological inventions.
The early internet empowered people, and we created some of our most amazing collaborative intellectual achievements like wikipedia.
Today, aside from being built by harvesting the IP of people, LLMs are shuttering our entire information economy. Sites I go to for info are either going behind paywalls or shutting down because they are being rapaciously harvested by bots.
Yes, history rhymes and repeats; however there are other tunes that have been played that sound similar but end differently.
To my mind, saying the AI will cause me to forget how to learn is equivalent to suggesting that going to school for instruction will cause me to forget how to learn.
I see plenty of “classically trained” people (as opposed to self-taught) that have no idea how to learn things on their own and at their pace.
I mean, in the era of youtube and the internet, any subreddit is filled with posts like ‘where do I start with X?’ or a endless rehashing of the same beginner questions from people that are not even able to use Google.
I am biased as a self-taught person that dropped out of high school, and it is baffling to see people have forgot how to learn things by themselves past 10 years old. There wouldn’t be so many people with student debts if that were the case.
Now like you said we're entering a world where anyone with a computer can pay a giant tech company thousands of dollars a year to spin up some agents for them. That's much less exciting to me, and I'm certain I would not enter the field if I were just starting out right now (assuming there even was a junior job available).
We've seen how big tech monopolies treat domains they control like search and social media. They try to extract all of the value, leaving nothing for the individual or common good, and they're quite effective at it. I'm not looking forward to them gatekeeping the field of software development as a whole.
but you can still do that, AI is not preventing you from doing any of that in any way.
People do want to learn and use new tech but instead what is promoted is an access to a proprietary and (increasingly more) expensive API.
Some suggested it could become web standard and I just hope it doesn't. React is beyond opinionated. It certainly has a raison d'être for some applications, but the problem is simply that it didn't put our less buggy or generally better sites.
Unfortunately, sama & co hunger for global domination makes them more and more expensive to run.
Meanwhile the front page is people complaining that using a particular word causes their evil genie to go haywire. You guys still call this stuff engineering? Writing requirements in prose, because programming languages are too hard? Fuck that, I’m out.
yeah i'm gonna call BS on that. what you describe was happening well before modern-day AI (LLM, agentic stuff etc) became mainstream: think of google accounts binding your identity to your searches, gmail, google adsense, facebook, instagram and twitter (and others).
And the products and services that do what you describe can do that just as well without ai.
So yeah the problem is absolutely real but AI is not the culprit here.
I kind of agree in general that it is a learned skill, but considering how unclear people generally are when they communicate, I'm guessing it'll take longer than a weekend to be able to catch up, especially catch up to people who've been working on precise and careful communication and language for years already in a professional environment.
the level of teaching involved would always mean the overall velocity of work slowed down.
some people say you can throw them the drudge work but i find that if you're doing coding right (e.g. you dont let your code base degenerate into a mess of boilerplate), there is barely any drudge work to do.
Me, too. But that doesn't mean I'm a great developer, just a shitty manager.
I suspect that will change sooner or later. Models will be cultivated over time the way we cultivate full-time employees now, with an acquired awareness of what they're building, new skills picked up in the process, and insight into how the larger system works.
There's a long-running instance of the model on the provider that's allocated to my organisation? Or are you thinking more of a server-side memory system, similar to the (currently very fallible) ones like Honcho and Mem0?
What happens when my org stops paying that provider? Do I get to take the now senior agent with me to the next provider? Does the provider have to delete it (and all that learning is lost forever)? Does that now become a free agent that can be hired by the next organisation like an employee (one that probably doesn't know how to keep industry secrets to itself)?
Karpathy's notion of a self-maintained wiki may suggest a direction for work in that area. As he points out ( https://gist.github.com/karpathy/442a6bf555914893e9891c11519... ) this is basically the latest of several attempts at implementing Vannevar Bush's "Memex" concept. I think it will find application in some form at organization-wide scope.
If the trend towards centralization holds, then yes, who gets to own and maintain the state in a stateful AI model is indeed going to be a big question. I hope it doesn't hold.
I just cant pretend that I'll get extra productivity while I'm training them.
Certain professions lend themselves well to a apprentice/master framework because the apprentice doing the drudge work that requires less skill frees up time. This can increase overall productivity while the apprentice is trained. Development isnt really like that though.
It is an investment, but it pays back for itself is my point.
I let the AI make decisions all the time. I often approve them, and I sometimes revert them. Most of the time they’re really good decisions based on my initial intent, but followed by analysis I didn’t make but agree with.
There's clearly some level where you want a human making decisions for even the most vibey of project, because without some kind of a spec about what you're trying to build and what features you want you'd get nonsense.
But like... maybe don't stress the details too much.
Yes, clearly. There was a meme out there, "just make something cool idk".
Statements like "Don't let AI make decisions" are made because of the loss of control we experience as mechanical parts of our work (such as writing to files) gets automated.
> A computer can never be held accountable
> Therefore a computer must never make a management decision > A computer can never be held accountable
> Therefore a computer must never make a management decision
If one has been reading a wide variety of books/papers/articles/whatever their whole life, and one has been mindful of how to communicate with the "written word" as it were, it takes about 3 hours to be wildly effective with this technology. I think it took longer to learn google-fu than it did to learn how to use this technology effectively.
That said - what we have learned in the last year could be compressed quite a lot - there are a lot steps we could skip, and 'learn by failure' that need not be repeated.
It takes a while to get the subtleties of it, it's among the most highly nuanced things we've ever encountered.
They're great at some stuff and terrible at other stuff in ways that are very hard to predict.
I'm figuring out new and better ways to use them in a daily basis, and I've been an almost daily user for nearly three years.
Let's say models can exactly and correctly write any code you ask of them.
- How do you break down a project into a sequence of requests to models?
- How can you most effectively parallelize the work - models will never be instant, so there will always be benefits in working out how best to use several agents at once
- Now that the models can handle the details of Lean, and Swift-UI, and Oracle stored procedures, and thousands of other technologies that you never got around to learning in the past... what can you do with those and how do you pick which projects to go after?
- How do you collaborate with other engineers and designers and product people in a world where you can churn out the right code reliably in a few minutes?
The models we have today are already effective enough to change the shape of our work as software engineers. As the models continue to improve figuring out and adapting to whatever that new shape is becomes even more complicated.
You need to be able to estimate how long an approach is likely to take in order to correctly prioritize your work.
Coding agents throw a wrench in this because some of the stuff that used to take a long time doesn't take a long time any more. I'm finding my 25+ years of experience in estimating software has been completely thrown off - it's taking a whole lot of work for me to start building up those intuitions about what's quick and what takes a long time.
With coding agents you need to think very carefully about how you design the agentic loop such that the agent has the right tools and information available to it to compete the goal.
I've been writing a lot more about that here: https://simonwillison.net/guides/agentic-engineering-pattern...
The difference is profound, and takes more than a couple of days to get your head around the implications. I'd summarise it as: "if you give a computer the same input it always produces the same output, but if you give a model the same input it always produces different output". Add to that the output is often wrong and it can't reliably follow instructions, and the difference is so great it breaks most of your intuitions.
The reward working with this piece of unreliable jelly is it can be far smarter than you (think the difference between a man with a shovel and a 20 ton excavator - they can literally find bugs in minutes that would take a human hours or days), and they know far more than you.
The engineering challenge is to make this near random machine produce a reliable product. It isn't easy.
The hype you see around them is it's trivially easy to get it to produce a feature rich but very unreliable product, as Anthropic demonstrates with their vibe coded claude-cli. I refuse to use it now. Among its other charms, it triggers a BSOD on windows: https://github.com/anthropics/claude-code/issues/30137 (Granted, it's just another Windows bug: https://learn.microsoft.com/en-ca/answers/questions/5814272/..., but if you are shipping to Windows you should be working around such bugs.)
I think this solidified an idea for me. A tool being smarter than me but inconsistent, is useless.
I can work with people who are smarter than me, because I can trust them, and I can trust them to own up or be held accountable for screw ups.
For a calculator, I can only hold myself accountable. However I cannot hold myself accountable for not knowing something I dont know.
I thought I could let this fly through keeper, but the idea keeps bugging me.
If the programmer does not vet and understand what the LLM has done, we call it vibe coding. If he takes personal responsibility for every line of code, we call it software engineering.
Lots of people use LLMs to vibe code (I do), and lots of people use them as a software engineering aid (I do that too).
The choice depends on the quality of code you need to produce. If an engineer is designing an aircraft wing, he would choose the highest grade of materials with known parameters. If he is building a kite for his son, an old newspaper would do in a pinch. Using 7075 Aluminium wouldn't just be overkill, it would mean his son doesn't get to see the fun his dad has building a flying machine in an afternoon.
A good engineer understands the properties of every tool at his disposal, and knows when to deploy each.
Disagree. It takes a lot of experimenting to find the right balance between sufficient guardrails and insane halluciations. And it'll be different depending on work domains.
I'm still refactoring AI workflows every week after more than a year or so and still working on it. Will probably be a perpetually ongoing effort as models change.
If you spend a year walking in circles, someone can easily close the gap with one step. Especially if models and harnesses are supposedly getting more powerful all the time.
The only real risk is that today there's an expectation from employers that you've got some AI experience under your belt you can articulate. But you can get that experience today.
You can still do creative thinking while using AI as a powerful tool at your disposal.
Some mathematicians like Terence Tao are comfortable doing this, for example.
Now the skill issue lies in whether your opinion is a good idea or not lol.
We also used to be much better remembering things, when we relied on oral histories, our memory skills have degraded quite a bit. And there's a quote from Socrates criticizing how writing is a crutch that degrades our skill (https://www.perseus.tufts.edu/hopper/text?doc=Perseus:text:1... , the last bit). Over time, we've just moved to valuing other things more.
That's extremely speculative, especially given there was a major event in 2020 which massively disrupted education worldwide.
I think it's just important context to keep in mind that these sorts of takes are very typical to top https://bearblog.dev/discover/ in the same way that certain types of posts are designed to rank well here. I considered migrating my blog there earlier this year and ended up deciding that, while I loved the product, the community was not healthy.
[1] https://forkingmad.blog/ai-summary-blog-post/
[2] https://blog.spu.io/you-dont-want-to-make-things-you-want-to...
[3] https://blog.happyfellow.dev/simulacrum-of-knowledge-work/
While waiting for Claude to finish, we talked about our hobbies outside of work, and the same guy will go into deep details on how steroids and the HPG axis works, and even gave me a spreadsheet with several NCBI PubMed links on the topic.
I think we are all naturally be more creative and opinionated in things we are interested in.
It remains the case that AI _erodes_ your ability to that.
So, eventually, after a few years, no, you can't.
Edit: meanwhile you're making yourself disposable. So, have fun with that.
I see no point in denying the technology, it's best to do what we humans do best: adapt with it.
I've read the article and to me it reads like a very angry rant, which is why I commented with something akin to "bro calm down"
Richard Stallman was unreasonable. So was Linus Torvalds. I'm hoping that something wonderful and entirely human-centered will come out of the anti-AI movement, up to and including a bifurcation of the internet.
Enforced how?
Take electricity adoption for example, with your adage then it means unreasonable pro-electricity people vs unreasonable anti-electricity people. We know how this turns out, I don't see a lot of people joining the Amish.
It's okay to have strong opinions (be "unreasonable"), but in the end humanity as a whole (the "reasonable" people) will judge whether your opinion is a good one or not. Only time will tell.
It will. That's why it's important to be a "Luddite" and to make our voices heard. Plenty are on the sidelines, plenty dislike AI, and plenty don't care to drink the kool aid and would rather exercise empathy towards fellow humans than welcome the era of the robotic overlords and their billionaire masters.
This place is an echo chamber and doesn't reflect the views of the vast majority.
I don't think it's hard to catch up if such a person changes their mind, though.
Some people who do use AI will be also left behind - those who use it to replace their skills without developing new ones themselves, and those who use it to do the same or worse work more cheaply. They will be left behind in a competitive world where others will work out how use it to do more or better work with no reduction in effort.
But if I'm to be expected to employ vibecoding in my day to day job as a software engineer, I'll dismantle my house and go live off grid somewhere in Alaska. I have enough power tools and knowledge to do it. Probably massively healthier for my kids.
I don't have any particular predictions going forward about it, but something I think about right now is, do I want to focus my time where the interesting decisions, the valuable contributions I make, are product-level thinking about what to build and what problems to solve? Or do I want to focus my time where the interesting decisions are technical ones, fully wrapping my head around a technical problem and coming up with a solution?
I do think both options are still available, and personally I love them both. But I don't know what types of coding would involve significant amounts of both activities anymore.
Product is when you’re seeing things as the one who have the problem and designing the solution in a way that is usable. Technical is when you shift to see how the solution can be implemented and then balancing tradeoffs (mostly costs in time and monetary resources).
While the code is valuable (as it is the solution). Building it is quite easy once you have a good knowledge on both side.
The issue with AI is not in their capabilities, but in people rushing to accept the first version when there are still unknowns in the project. And then, changes costs almost as much as redoing the project properly.
It sounds more like there is no chance that most of those people will stay employed, regardless of how "ahead" they try to stay.
- https://publichealthpolicyjournal.com/mit-study-finds-artifi...
But it's definitely possible to use AI without letting it think for you. OP should at least acknowledge that.
Those who dogmatically refuse AI outright may be disadvantaged for some things in the future. But it's also probably hyperbolic to say they will be "left behind".
I've seen some produce stuff without really understanding it, barely review anything, and pretty much suffer from imposter syndrome.
Obviously people who are motivated by curiousity will have a different view and those who value creativity will end up thinking otherwise.
Also, it's basically impossible to separate the technical capabilities with the big money fascists pushing it.
Because it doesn't make sense to be better than a tool. A woodworker could use a hand saw and take an hour to cut wood... Or he could use a buzz saw and cut it in a few minutes. Is the woodworker any less of a woodworker when he uses a buzzsaw vs a hand saw?
Outsourcing thinking to AI is not healthy, and certainly if everyone used AI like this we're doomed.
I still think it's true that those who don't use AI will be left behind, but it's a bit tautalogical because the thing they're behind left behind on is AI. A lot of the biggest companies on earth are putting a lot of money in AI, but if you're OK with working for a company that is not putting all their money in AI that's perfectly fine.
Just like block chain was everywhere ten years ago and now is just kinda _there_. If you got in before the hype you could have made a lot of money. If you didn't, you were left behind. I was left behind and I'm OK with that.
Just as many people leading sedentary lifestyles have to make a deliberate effort to exercise, because inactivity is really bad for our bodies, I think we're going to realise that a similar process is necessary for our minds.
You really want to be spending a bit of time every day operating at your cognitive limits - trying to fully engage your System 2 - if you want to avoid brain atrophy. Coding used to kind of give you this exercise for free, but you can go really far with just your System 1 nowadays - literally get things done while scrolling Reddit.
I'm trying to allocate 30-60 minutes a day to doing something difficult, like writing code by hand for an unfamiliar problem or reading and summarising difficult papers without AI.
But as a programmer I am quite baffled when peers my age, often without kids struggle in this kind of way. My essentials are rent, food, metro card, library card. When I hear people who make what I make say that living requires a lot of money that usually includes two dozen subscriptions, a few grand in useless electronics per year and ordering food most of the week.
Personally, I have pets, a partner, and thoughtfully selected and meaningful possessions. I don't collect crap, and nothing I own or do is particularly extravagant. But I'm not exactly living an ascetic life either. A pretty typical lifestyle, I'd say. And I don't consider any of this a moral failure, if that's what you're getting at. Though admittedly it's not as affordable as living like a monk.
I'm very much in favour of participating in culture, real culture though. You can ditch the 1k Taylor Swift tickets, Disneyland visits, and high end gym for the 20 bucks local jazz festival.
I think the people who live socially like ascetic monks are ironically the people who build themselves a home gym for 10k and then complain about not having any time for friends because everything is too expensive
Anyway..
https://news.ycombinator.com/item?id=47932937
Left behind with my money,
https://news.ycombinator.com/item?id=47933355
Left behind with my intact data,
https://news.ycombinator.com/item?id=47911524
Oh, the horror. I am being left behind.
But "your dignity"? You mean like "I feel shameful over that people saw that my writing was actually AI?" or something else?
Well if you don't have dignity in the first place, its hard to have any shame over losing it
If it can be done by AI, then you have no hope of competing with the quantity of AI output that anyone can trigger in very little time.
As my job seems pretty secure, I can ignore AI for as long as I like.
It is a tool just like syntax highlighting, code completion and refactoring tools before it. You need to know how to use them, where their usefulness ends and you should probably have an idea how to do it yourself without the tool. It is okay if you will be less efficient, but it's bad if you just can't.
A bimodal strategy gets you the best of both worlds: the ability to rapidly explore and develop ideas, and the ability to critically and cautiously assess them.
Diversity fuels evolution.
This AI/LLM push from leadership is so damn tone deaf, like "you better do this", "ai layoffs", etc. I feel like they are jumping way too hard and fast into the "post-employee" thinking and deserve every bit of scorn from laymen.
It’s gonna be painful and many firms will go bust - not because of being left behind. But they got so deep in pushing llm’s their competitors came in and offered their customers what they wanted - the customers don’t care how you produce it. They want the thing to that leaves them in the best economical state. And this llm mania is gonna cause many firms to forget this and go down paths they will later regret.
The author chose to take offense by connecting with a false dichotomy presented vaguely in a way that serves no purpose other than dividing and poorly labeling everyone in an area where much nuance applies.
I think this is perhaps a side effect of consuming too much content and feeling overwhelmed with it.
Engaging with stuff like this only amplifies its effects, how about do anything else instead? Maybe learn something new, like how to channel your anger.
Why spend your life "learning" something whose whole deal is about not needing to learn? Even if you gamble incorrectly, its not going to be hard to get into!
Like, what, if I don't start practicing now I am not going to be able to... express concepts with natural language as well?
Using AI is and "think, how to write, how to do a simple reliable search, how to tell fact from fiction" are not mutually exclusive.
Yeah I don't think this line of reasoning is to be taken any seriously at all. It can be... left behind.
Same. I don't really care about productivity or if AI is so much more productive, tbh. I'd rather just change careers at this point. I'd prefer not to just be a full time code reviewer while my agents go do the actual work.
But I'm also tired of this in between state. Either rip the bandaid off already, fire everyone, and force governments to implement UBI so I can finally be free, or finally admit that the productivity gains have been vastly oversold and the LLM apocalypse is only a half truth, half grift and get on with our lives.
I'm also tired. I wish this would happen too.
I just don't think it will because it would devastate the market.
Well of course too much is bad for you, that's what "too much" means you blithering twat. If you had too much water it would be bad for you, wouldn't it? "Too much" precisely means that quantity which is excessive, that's what it means. Could you ever say "too much water is good for you"? I mean if it's too much it's too much. Too much of anything is too much. Obviously. Jesus.
And that’s exactly what happened to those who started using an abacus, a calculator, or a computer.
No, it is not an issue that they might „forget“ how to search for the information in a book from a public library.
No, it is not an issue that they might „forget“ how to get this information from a book in a public library.
Sure, he'll get it done twice as fast and you might notice some tricks as you look over his shoulder. But when you need a second door hung, you'll either have to start learning from scratch or call him again.
Everyone seems to know you can't trust the AI output, and that it is on you to review it. But whenever I talk to people who claim to be getting big benefits, there is always a moment they reveal that they are not really reviewing the output. They are just going with it.
Similarly, so many who claim to use AI as a search index eventually seem to just trust the summary instead of checking the references to figure out whether it is regurgitating fact or fiction.
I don't really know if these users always had low quality standards or low diligence, or whether the tool usage degrades them. But I see the correlation among the friends-of-friends network I can observe.
Nevertheless, the responsibility of whatever a human produces with AI is still on the human.
With that said, knowing how to use AI the way it's right for you can give you a huge advantage. You don't have to though. And there is not a standard way of doing it.
What I recommend to everyone is give it a try and see if and how it could help you. At the end, you have to make the decision based on your constraints and what you're aiming to and can sacrifice, including but not limited to speed, accuracy, learning, etc.
To me the idea that a GPU which costs as much as a car must read its entire VRAM just to output a word sounds incredibly wasteful. I'm exaggerating here, but it is literally reading gigabytes of data and processing it to produce relatively little information.
Some data is truly worth the effort, but the majority won't be able to afford this long term - especially when those who capture the market increase prices.
Kind of how I feel about Bitcoin at this point. The coins take so incredibly long to mine if you aren't in a pool that it could cost hundreds of dollars in electricity to own a fraction of the coin months later.
> [...] they'll forget how to fucking LEARN. I think that's the part that makes me the saddest. What a beautiful thing it is just to learn stuff.
I love learning. My life of self-education is so much richer with LLMs to help me.
There are dozens of other arguments for not engaging with AI. If your reason is "I love learning" I recommend at least dipping your toes in before you declare that AI is a hindrance, not a help, to people who love to learn new things.
"Chess players who don't use engines will be left behind", they say. I can't emphasize enough how much I hate it when I hear/read shit like that because I'm pretty sure, in fact, that what will happen is the exact opposite.
People who rely on engines are the ones who will be left behind. They'll forget how to think, how to move the pieces, how to solve a simple straightforward mate in 3, how to tell victory from stalemate... they'll forget how to fucking LEARN. I think that's the part that makes me the saddest. What a beautiful thing it is just to play chess.
If you think Deep Blue can do better than you, why would you just let it? Why wouldn't you aim to be better, to learn how to be or do something that a chess computer would never do?
Also, the world isn't as trivially solved by computation as a game of chess, so maybe delegating your job or how to be a better human to ChatGPT isn't as much of a winning strategy as getting the computer to suggest chess moves.
Deeper reasoning, longer term planning, and more efficient solutions have always separated amateurs from experts. That experience cannot be applied asynchronously or reduced to supervision. It has to be "in the loop" and there is always a lot of out-of-band information that only an experienced eye would notice and can't be trained into a model.
Maybe there's a way AI can be used to make developers better but it mostly just seems to be the equivalent of grand masters saying how great vibe playing is because now they can play 1000x more games every day. But don't worry, they're still steering the games.
Unfortunately this is absolutely true for classical chess at the professional level, w.r.t. preparation.
Not detracting from your point though, for the other 99.9% of chess players.
And I'm sorry to nitpick - but "People who rely on AI are the ones who will be left behind" is NOT the opposite of "People who don't use AI will be left behind".
Today, we are hearing a similar claim: “If you can describe the program in natural language, programming is basically finished.” But the industry is now discovering that describing the program well is the hard part.
This is also why ideas like harness engineering are appearing: methods for controlling the range of outputs, from poor to excellent, that can emerge from minimal input.
And honestly, I do not think the “vibe coding” phenomenon is entirely bad. The essence of programming is automation. Many people were previously limited because they did not know programming languages. Now, through AI, they can express themselves and turn that expression into working apps. Seeing this, I understand how deeply people have wanted to create.
I write industrial software that runs in large factory environments, and because of the nature of my work, it is difficult for me to use AI directly. These environments are usually closed networks, so AI does not really benefit my own production work. Even so, I still defend AI, because it functions as a new kind of voice that allows more people to express themselves..
Of course, capitalism distorts this. Many people use AI to chase money and capital, and as a result, a lot of low-quality apps are being produced. But on the other hand, what is wrong with the motivation of wanting to make something one wants to make?
I have been studying the history of programming, and I like Dijkstra’s famous line:
> Computer science is no more about computers than astronomy is about telescopes..
To me, this means that computing is fundamentally about automation.
AI has existed as a research topic almost since the birth of computers. We tend to think of it as recent, but it is a field with a history of more than sixty years. Starting from early work such as the Perceptron, there have always been people claiming that AI was a fraud or an illusion.
But now a new seed has germinated. The amount of complexity that a single human can handle has increased. Historically, the techniques for managing that complexity were things like programming patterns and software architecture. And even people who strongly argued for software architecture also warned that if architecture becomes detached from code, then something has gone wrong.
Memes always damage the essence of ideas. As information circulates, it degrades, and eventually the original meaning disappears.
The Dunning-Kruger effect is a good example. The original paper was not simply saying, “ignorant people show off, while knowledgeable people do not.” It was more about how both less competent and more competent people can have difficulty accurately assessing their own metacognition. But the idea became distorted.
The same thing happens to many famous ideas in programming. Knuth’s statement about premature optimization is also constantly distorted as it circulates.
In that situation, can we really say it is always bad to step away from online communities and learn through AI while cross-checking against books?
When I see people making extreme claims about this, I sometimes find it absurd. Of course, many people may flag or downvote my comment. But this is how I see it.
If you didn't embrace OOP Test driven development Behavior driven development Events driven development Pants in head driven development SOLID DRY Cloud first Virtualization everything Microsservices Serverless Everything js Everything ts Everything Microsoft
This will never stop.
You either let someone be in the middle of you and what you want to accomplish, or you will be left behind.
Think about the most mediocre person you know. Now remember 50% of people around you is dumber than that
People who drive are in worse shape than people who walk
it's clear that LLMs are unique in that you actually do have the capability to turn your brain off and blindly trust whatever it does for you. but it should be equally clear that that's a stupid approach. people will still use their minds, and this use gets empowered with proper use of LLMs. it's that simple. ffs, we take the fact that they pass the Turing Test routinely for granted now. let's not forget that this technology is legitimately incredible. it stands to reason that you are seriously handicapping yourself by not trying to use it.
There is absolutely not doubt; and it will be impossible to avoid as using 'plastic' or 'electricity'.
The narrow challenges of 'AI aided development' or 'AI aided creative work' are legitimate - that part is real and fair, but it'd be an over-statement to contemplate 'not using it'.
The cyclists who keep their muscles strong the 'hard way' ... will win the delivery war vs. cars!?
The carpenter who hammers every nail and saws every plank by hand 'the hard way' ... will win over the guys using power saws and nail guns!?
No - AI is changing the landscape.
What is 'hard and easy' are changing.
We won't need some skills, we will need others.
It maybe harder to maintain some critical skills, but the upside is obvious.
What is fundamentally missing from this treatise is that 'there is always a hard way'.
Personally - I have never been more 'cognitively overloaded' than ever. The AI 'amplifies' the depth of complexity one can reach, it's just at 1/2 a layer of abstraction above the code.
Driving a 'race car' at the highest speeds - is as challenging - and perhaps more so - than riding a horse.
The 'instinct to push back' is fair and there are innumerable legit criticisms ...
... but AI is just a new part of the stack and it will be as horizontally applied as 'software or the transistor' - it's not reasonable to think one could or should avoid it entirely.
with AI agents, you're obtaining a mildly lossy perspective of the code itself. whereas if you wrote it by hand, you'd have a more concrete understanding.
This is not too different from an engineering manager directing junior developers.
The stereotype of the engineering manager who forgot to write a line of code is not wrong.
i - AI is going to interject in so many things and so many ways beyond 'helping you write some modules' so consider that.
ii - AI 2 years ago was useless for code, you can see how well it works not, and this progress is still very real. By this time next year, the power will be more evident, making the position harder to take.
iii - to your point - the real answer is 'abstractions'. We used to write machine code by hand as well, until someone came up with FORTRAN and C etc. Now, people have 'forgotten' how to do that, largely, because we don't need people to do it.
AI is crudely that abstraction. You don't have to know a lot about some things.
Now - it's very fair to highlight the fact that the abstraction isn't very clean (!!!) but that will come over time.
So yes - for writing software today, we're '1/2 a layer abstraction up' - and it's 100% essential to keep an eye on the code, the architecture etc. - it's 'not fully there' but it's better to look at this through the lens of growing capabilities because over the horizon, the argument starts to tilt.
It's so completely obvious, that anyone denying it has to be living in some kind rhetorical bubble.
It's truly a feature of 'online rhetoric' like HN/Reddit where people can consider these asymptotic postures and take themselves seriously.
We will use AI like you use plastics, cars, electricity, computers etc..
That's it.
I'm sure there were a few people who thought that 'hand writing machine instructions' was the 'one true way' of writing software, but hey, what would we call them in hindsight?
There are so many legitimate ways to be curmudgeon or wary of AI, but this reactionary stuff is anti-reason. It's not an argument, it's guttural, we should just ignore it.
People not using AI will be about as useful as those refusing to use e-mail or computers.
It's absurd.
AI may replace some cognitive activity, it also required cognitive intelligence to use 'slide rules' - which have been replaced and we have not looked bad.
It's not a bad rhetorical question - but it's moot in the face of the question of 'should we use it or not'.
It will do a lot of things for us - that part is inevitable and unavoidable.
We'll have plenty to think about.
If it's good, lots of people will use it commercially. If it's generationally good, everybody will use it commercially because commercial use is about competition. It either gets banned outright, like steroids, or — if it doesn't get banned — those who use it will have a clear advantage and that will lead to a very small number of people who don't use it (in business).
This is not really something that opinions are required for because if you think LLMs are going away, your opinion is historically incorrect. Things that reduce toil and increase output do not go away.