I feel the same way. I used to love to get into a flow state while programming and feel very proud of clever code and elegant architecture. But now LLMs are so much faster it feels somewhat pointless.
The code is still there, you can still flow with it. And your ability to do so (along with the LLM) will only produce superior results in the long run.
It's just, yes, like all of us .. the temptation is there to not do that.
Deletionists have been circling html since the beginning and have only managed to get rid of document.writeln() they tried really hard to get rid of framesets but ultimately failed.
I suspect in the long run a similar lack of "success" getting rid of Apache.
There are probably other things that are here to stay.
Personally, while I'm becoming increasingly convinced there's going to be a wave of SWE layoffs, another part of me doesn't grasp it yet. I don't feel it.
I was an AI skeptic for a while but it's literally doing my job for me right now. Every month it needs less of my help.
Once they replace SWEs, they're going to turn around and replace every other knowledge worker yet. But most people haven't processed it yet.
The AI field has known, but repeatedly ignored, that the knowledge work group would end up first on the chopping block, especially for those where everything they touch is digitally encoded already.
I used to joke that we're all really in trouble when robots can sort out your socks, but as of about a year ago I started seeing disturbing progress on that incredibly gnarly problem.
Most people really are going to be in denial about this as a coping mechanism, but the truth is the worst thing would be slowing down the spread of this through the economy creating large minorities of disenfranchised. It would be far better to get to the end asap, whatever it proves to be.
I don't think it was ever ignored that knowledge workers would be the first affected by AI. I remember hearing for a long time that lawyers, doctors, accountants, etc. would all be obsoleted by AI.
I do think many (including me) were surprised that software engineering would be the first kind of knowledge work to be successfully automated. We assumed it would be one of the last, because we'd be the industry using AI to replace everyone else, not the industry using AI to replace ourselves.
> the truth is the worst thing would be slowing down the spread of this through the economy creating large minorities of disenfranchised. It would be far better to get to the end asap, whatever it proves to be
Getting to the end asap is absolutely going to be "the vast majority of people are disenfranchised", how could it possibly be anything different? The only lever most people have is their labour, and that is being stripped from us.
Even if we aren't impoverished in the future due to some UBI scheme or whatever, it's hard to imagine what that looks like other than "disenfranchised". How do you have any kind of clout inside a system that you are fully dependent on?
That's called "elite overproduction". Too many people going to college when the jobs aren't there. In the US, about half of college graduates have jobs that don't require a college degree.
It's much worse in less developed countries. The article today about essays in Kenya mentions that the country is creating far more educated people than it can use. Egypt had a scheme where the government guaranteed a government job to everyone who graduated from college. They had to give that up when the oil ran out.
It's hitting top people first, which was unexpected. Mathematicians who've won the Fields Medal are worried, and some are giving up math.[1]
I've been worried about AI replacing programmers since the release of ChatGPT 3.5. And I've been consistently surprised through all of that time that people didn't see it coming.
For example, from early February 2025:
> I'm surprised to see a huge disconnect between how I perceive things and the vast majority of comments here.
> AI is obviously not good enough to replace programmers today. But I'm worried that it will get much better at real-world programming tasks within years or months. If you follow AI closely, how can you be dismissive of this threat? OpenAI will probably release a reasoning-based software engineering agent this year.
> We have a system that is similar to top humans at competitive programming. This wasn't true 1 year ago. Who knows what will happen in 1 year.
Overpaid techies learning they’re still just as much a means to the capital class’ ends as the factory workers whose jobs the US shipped to lower wage countries.
Class consciousness y’all. You’re smart enough to Marx and Piketty so do it.
I'm sitting here with DeepSeek 4.1 Flash, paying pennies to hammer out on projects that I have wanted to do for years but no employer would pay me to do. And should those pennies become difficult to spend or the API taken away from me, I have a DGX Spark on my desk I can run local models on to get the same thing done, just slower.
It's somewhat the opposite of the concentration of the ownership of the means of production that the old bearded man spoke of. (And I do like my Marx, honestly).
That said, yes, this does have the effect of potentially devaluing the work generally and droppings its compensation rates. Though I'm not sure we've hit that yet. So far places I've been working just pay me the same to get more done.
I was just having this conversation with someone. It's less sadness for me and more dread though. All of the sudden, everyone's technical expertise and the things that they spent their life getting good at, are accessible by everyone.
I don't know if this makes sense as a society. Say whatever about what it means ethically, legally, etc., but I'm not sure this kind of thing is sustainable in the long run for a functioning society.
I did well in school and the praise I received for excelling in my math degree did shape my life and impact my motivations. I do like praise.
But outside of social reasons, it's understanding I crave. Thoroughly understanding a topic is awesome. I remember the feeling of acing my qualifier exam in grad school.
Those LLMs change nothing fundamentally about this motivation and need and capability. I can't type a prompt, hazily skim the wall of text, and come to comprehensive understanding. I still have to work.
What I like about LLMs is that they can increase my understanding by helping me learn more efficiently. What I don't like is the temptation to turn off my brain.
Well yeah it conditions you to turn towards it anytime you hit a roadblock.
It is in a precarious state right now where it is too good not to leverage but will bite you if you get sloppy.
I still write things by hand sometimes because it is fun, but it does make you feel stupid when you spend 2 hours fixing something a 2 min prompt could have fixed.
My son told me yesterday he wants to program video games for a living. He's 13. He's looking at how it used to be done and seeing the pride people had in doing it. He hasn't seen how it is now.
Not sure I have the heart to tell him how terrible the industry has become. Feels like telling a kid 30 years ago that being an astronaut is pointless because they no longer go to the moon.
You are looking at this all wrong. Give him the tools now and watch him be a better game dev today than a professional engineer from 4 years ago. He can work on his dreams as of this exact moment.
I was the same at that age (decades ago, but still). And I'm kind of glad I didn't go down that path, seeing friends and colleagues who did. There's a (I think) Folding Ideas video where he's talking to some 13-yo Roblox developer who already has that jaded veteran kind of vibe to them. Which might be the most salient visual. The thing is separating the consuming experience from the creating experience, but also recognizing the validity and importance of both. It's obvious but hard to really internalize.
A good parallel I've discovered as an adult is wine. I love it, and collect it, and can talk at length about terroir, techniques, etc. About ten years ago I wanted to get hands-on experience, and thru a friend was able to intern at a local winery during harvest.
That was one of the most intense months of my life, and I was in physical pain for months afterwards, even as a relatively fit person. So much of that elegant, philosophical final product is manual labor, lifting lots of heavy shit, and endless repetitive motion. It was fun, but it was also brutal.
It left me with even more respect for the people who dedicate their lives to it. But also the realization that I'm much more comfortable being on the other end of it. And that's not only ok but absolutely essential, because the producers need people to buy (and maybe wax philosophical) about their product.
So yeah, by all means give it a shot. But also I think it's important to communicate that there are many valid and satisfying ways to engage with something. Consumer/maker isn't a binary.
The thing you love and made your identity was always the cruft middleman of getting from one place to another.
People were sad when we moved from horses to cars; they loved their horses, they named them. That didn't stop the slow, glacial, detached march of the calendar, of progress, and in this example, of the wheel.
Excellent post. How many careers and industries has software killed? Now it’s our turn. We sold out for crypto scams and toxic social media and surveillance capitalism. We deserve zero sympathy.
This strikes a chord with me, though I came to a somewhat different endpoint, at least for now.
My perspective (so far) is that it was probably never a good thing for me to tie up my identity/sense of self in my ability to write moderately clever and generally mediocre code.
I do still enjoy writing my moderately clever and generally mediocre code, and for the sake of my sanity, I'm trying to do more of it.
But either way, I don't think it was never a good viewpoint to make me self-conception pivot around my ability to be useful by writing code, and maybe it's a good thing that LLMs have crushed my ego on that front.
Relatable. I've taken advantage, in my career, of my aptitude to learn how moderately-complex systems are put together and my problem-solving instincts which are honestly nothing genius level. Just methodical and grounded. Someone created a bot which takes a bug report, checks logs, the database, and the code, and writes a write-up of what the problem is, complete with how to confirm or refute the hypothesis, and a description of what would need to be done to fix it (including any data fixes). Following these plans is very effective 90% of the time. Now, as a manager, I have to hurry up and get better at a lot of other things because in my opinion a lot of why I was usually kept around was because I was so good at knowing how things worked well enough to explain things like that, or to help validate new tech design. Plus I have to learn how to use AI effectively myself.
Came here to share the same general idea about using this as an opportunity to recognize when we've based our sense of self-worth on something external, and therefore out of our control and unstable. Which I also hope wouldn't come across as patronizing, though I can see how popping into the comments to say "I hear you're sad, but it's actually because you've failed at understanding yourself!" is far from seeming helpful lol. But the author uses a lot of language that directly signals this.
I'm navigating the same feelings as the author, though maybe to a lesser degree partially because I had to kind of reckon with this last year due to being laid off and burnt out. The thing that's helped the most by FAR is challenging myself to reframe my sense of self around things that being a Software Engineer was a mere _manifestation_ of, like being curious, creative, going deep on technical problems, trying to optimize things, etc. There are infinite ways to embody those tenets, and it helps to make the changes to our industry feel less existential without requiring me to feel like a different person, someone without a place.
Kudos to the author for being so vulnerable. The video presentation format is quite intimate, and a very different feeling than you'd get from a youtube video.
This is basically me in 2026. Life has no meaning and I'm waiting for my money to run out before I end it. Hopefully I can find a job that doesn't bore me but I'm not positive.
We're much more than any economic system. The sun still shines. There's a lot out in the world to see.
Even in the worst case scenario, that we're all jobless. There are mountains to explore. Beaches to walk. Forests to camp in.
I don't know where this all is headed. But, I do know each human has value. Your value is far greater than any job, skill, or the number of green pieces of paper you have.
I have to admit, as a former JW, the linked article is actually not proselytizing, instead it's actually got worthwhile advice.
We are each here for a reason and we all touch many lives. We may not ever know, while we're on this earth, many of the important roles we play, just by being ourselves.
I'll add. I once got stopped by a bunch of JW. And I wanted to practice my Dutch, and they were willing to talk to me in Dutch.
They showed me a video about depression. And it was actually exceptionally good advice. It spoke about speaking to a therapist. There was a small bit about prayer, which was obviously not my bag. But I was pleasantly surprised at how grounded the advice was.
I often feel like this as well. It's hard to talk about it with people, because they panic and immediately start sending you literature about suicidality and giving you hotlines, etc, etc, as if I haven't already seen all of that before.
I'm not actively suicidal or anything (and I know all the signs of risk, and what to do about it), but there's an underlying feeling that life just hurts and has no meaning. Currently the thing that helps me stick around is the idea of "doing it for the plot". Life ends in short order either way, so I figure I'll just stick around and keep it interesting.
> I'm waiting for my money to run out before I end it
Please don't do that.
Someone emailed us a link to your comment and was concerned that you were in this level of distress. I didn't have another way of getting in touch with you, so I thought I'd post here and say that there are strangers on HN who care. Perhaps they (or we, rather - I'll include myself too) can't do much beyond that, but I hope it makes at least a small difference.
I’ve been through some really bad stuff, and one thing I learned the hard way is this: don’t automatically trust your feelings or your thoughts.
Your brain can come up with completely random, untrue conclusions and make them feel 100% true. It can convince you, beyond any doubt, that something terrible is going to happen in the future. When in reality it’s just fear making predictions that never come true.
So hang in there. Use your logical brain. Ask yourself: What actual evidence do I have that what I’m thinking is true? Separate what you know from what you fear.
Then keep moving forward. Push through the fear. Let time pass. Give reality a chance to prove your predictions wrong.
Because often the thing that feels absolutely certain is nothing more than your brain telling you a completely and utterly random story. And you don’t have to believe it.
Hey, op here. Not sure if you're serious but please don't end it. My message to myself and to you is that your value is not in your work or your abilities, but inherent in yourself. You are unique and amazing. Please look around you for people you can trust and talk to them. Get help if you need it.
It’s been hard for me to overcome the feeling that anything I can do, the machine can do better and faster.
I know people have been saying “If you’re worried about it replacing you, you weren’t a good engineer to start with,” but it doesn’t make it hurt any less. I had a lot of fun solving problems and thinking deeply, but now my time is much better spent reviewing the plans and code of a machine that’s done that part for me.
Yes it’s still a hobby I can do on the side, but at my day job it’s just sucked all the energy out of me.
Unless you were literally the best engineer on the planet, it was always the case that someone could do what you do better and faster. Now, we're either at or approaching a world where even the best engineer on the planet has the same reaction as you. The playing field has been reset. The question for your future is which of you can make better use of the genius engineer in your pocket. Perhaps that involves a different set of tasks than you imagined yourself doing, but fundamentally, all good knowledge work is just creative problem solving. If you were good at that at one level of abstraction, I would bet you will be good at it at another.
Llm is a tool. Programming had used lots of tools through the history. Programming itself, is a tool for societal change. Building identity around it is ridiculous, I believe much healthier identity building is around what you can do with programming, rather about the fact of programming. Just my two cents
> I think I've felt sad because of the disrespect.
How is this to do with AI? When were we ever respected as an profession? My career has given me considerable experience of non-technical people in senior positions having limited respect for those with technical skills and running roughshod over their interests. The exceptions to this rule were mostly either theatre or a fortunate consequence of people like us getting into those senior roles. That's not even beginning to count all the masses incapable of perceiving us. Consider the outpouring of hatred for AI over artists and the scant sympathy for any of us.
The sorrow is mostly the status quo returning in typical money driven orgs. If you read stories of the 80s then technical staff were incredibly poorly treated and paid and those voices will rise again, empowered by their hubris and slop.
The only positive is now its easier to replicate the conditions of those orgs as you don't need to shell out over $100k per annum for a dev, given that these tools are considerably cheaper. This means we could compete with those forces without needing their pockets. We could even collaborate and somewhat destroy all those companies.
However we wont because we're more interested in little puzzle boxes and arguing at great length about what something should be called until finally agreeing upon one of the very worst options.
Good essay. The author seems to recognize his hypocrisy and why there won't be a lot of sympathy from society.
As someone who wrote his first program in 1980, I've actually been really excited about AI. This morning I already wrote two web apps for my kids. I was never a web developer, so I most of the dev work I did was in the service of backends or low level stuff -- but I rarely ever connected it to anything I could use day to day.
Now I write programs daily that I use. An example from this morning, my son is working through a workbook, but the answers are all on one page, but its hard to look at an answer and not see other answers. So I made a simple web page where you give the problem number and it gives you the answer. Super simple, but ended up being exactly what he needed. I'm finding that I'm just crafting up these 5 minute to 60 minute projects several times per day. Things where in the past I'd just say, "too bad" -- now I'm like... "hmm... what does a solution to this look like?!"
So I'm not writing much code any more, but I don't know if I'm having any less fun.
I enjoy working on personal projects, but I have never enjoyed coding in professional contexts, so AI doesn't actually reduce my level of enjoyment at work at all. It actually increases my level of enjoyment, since it does a good job of handling what to me are the most boring aspects of professional coding: writing boilerplate, remembering nuances of language syntax and APIs, etc... and allows me to focus on what to me is more interesting: architecture, ideas, business strategy, etc.
1. It is clear that many people, including the person in the video, has strong identities tied to a certain skill that they have spent years honing and perfecting. The sudden loss of agency and democratization can be confusing, disorienting and frankly demotivating. We need to acknowledge that.
2. Unfortunately, this has played out several hundreds of times over the years in the world. I'll take an extreme example to illustrate the point. Movie stars. There was well trodden path of getting to stardom and adulation, and everthing that comes with it. You move to LA, you work on your acting skills, you take your body aesthetics seiously and with plenty of luck, you might get cast in small roles before you work your way upto stardom. And the riches that come with it, cars, fans, brand deals .. great. And then came internet and YouTube where a guy in basement can garner millions of views, brand deals, fans in millions with little to no effort.
This is the same pattern with an accomplished musician (think a drummer) and someone who uses a drum machine software to drag and drop the "amen loop".
Exact same pattern with broadcasters and internet-journalists.
Yes it is demotivating, yes it is confusing but such is progress. Adapt or die. Its ruthless.
1. Corporations misappropriating $9Tn of opensource human works to sell isomorphic plagiarism tokens is not progress, but rather an unsustainable ectoparasite people will not tolerate very long.
2. Indeed, the invention of the Cotton gin was credited for prolonging the economics of US Slavery for another century. Those folks were probably upset for a good reason.
It is a self correcting problem, as anything that could be scraped has already been stolen. Only the chat user data remains, and LLM are causing a 17% decline in core skills.
Note too, there is a trivial counter-proof to recursive self-improvident theories that any child with calculus II can solve. No I won't warn them this time, as we are waiting to strip the data-centers for discount infrastructure.
The market correction correlation with Shrek movie releases is undeniable, and June 2027 is closer than you think. =3
> 1. Corporations misappropriating $9Tn of opensource human works to sell isomorphic plagiarism tokens is not progress, but rather an unsustainable ectoparasite people will not tolerate very long.
I think about this a lot when I look at the roads, freeways and railroads of the world. In pursuit for building it, billions of trees have been cut down, land fundamentally shifted, rivers cut through, and wildlife uprooted.
Yet, roads, freeways and railroads allow for growth. Someone facing medical emergency can reach a hospital on road because it exists. I get to eat fruits in winter because it is imported and transported to me from warmer places and/or storage units.
Human progress and prosperity comes at a cost. Nothing in the world is free (perhaps maybe sunshine?) and the life that you're yearning pre-AI is also extremely tainted. Infact, you'll be hard pressed to find an era when human meddling hasnt had bad impact on the world (right from farming and domesticating animals).
I ran the numbers on Sam Altman flippant statement about wasted energy, and found roughly 40000 human 20 year old lives worth of energy are exchanged with every frontier model training-cycle release at OpenAI.
Some history... Westinghouse Electric sold turbines to power stations in the 1950s, and promoted electronic home appliances to drive energy consumption to close the sales loop. While the scheme was not a direct business success, for society it improved just about every Americans living standards.
The problem is LLM cost resources, broke that loop, and only bids down the quality of life people can afford. You pay more, have less, and reduce taxable labor incomes into demand deficient classes.
This will not end well for people even if LLM are recognized as mostly hype. =3
> You pay more, have less, and reduce taxable labor incomes into demand deficient classes.
This will not end well for people even if LLM are recognized as mostly hype. =3
Depends on how you view efficient market hypothesis. Computers and Memory was prohibitively expensive. "Putting a computer in every one's desk" was literal rallying cry for Microsoft because it appeared audacious, just like saying "Unlimited LLM usage for every man, woman and child" or some such.
Will it happen? Nope, but if the technology is useful the markets it'll spur competition and prices will come down. Heck, we dig plant juice out of the earth from deep sea oil rigs so that I can fill up my car to go walmart to buy soda. Pretty preposterous when you think about the closed loop.
>if the technology is useful the markets it'll spur competition and prices will come down.
Regulatory-captured markets have nothing to do with free competition, or bidding down price pressures. You do have monopolistic competitive firms drawing $4Tn in loans to artificially hyper-scale demands for NVIDIA hardware that earns you -$2.50 for every $1 in revenue.
Your home heating bill may double this winter because the economic engine is currently post-consumer/broken. We shall see how this plays out. =3
stop calling them AI... those are LLMs so just statistical prediction software trained over other coders code... if no one makes code anymore then LLMs will learn from their own written code that statistically contains errors that will propagate and enter the feedback loop amplifying...
The loss of identity is quite destabilizing, and our industry is not doing enough to address it. For example, one thing we could do is move identity to value to the user, in a structured way. It is possible, but hardly any team is thinking about it.
I feel the opposite of sad, I am very very excited. I love programming and I love technology, but the stuff I'm doing now is beyond what I could not do on my own, and now I'm spending a few hours building complete end-to-end stuff. This is the golden age for builders.
I created a fully working house simulator, I made a math test creator for my kids, I fixed up some other side projects that I had and added observability, as well as installed Prometheus and Grafana and built grafana dashboards all using AI. It's wonderful.
I plan on building real-time stock price monitors with personalized alerting, taking images from my security cameras and making time lapse videos, trying to do real time predictions on football games/individual plays by having AI monitor the game and based on previous history predict plays/outcomes specifically for sports betting. I need to find the time to figure out how to do all this stuff but what used to be a pipe dream is now 100% possible and accomplishable within days.
This is exactly how I feel. I can roughly understand why people feel sad, but I cannot relate. I feel more capable than ever before and am simply excited. LLMs have been a great learned tool for me especially to venture into new programming domains in my personal projects.
For me, I've honestly never been much of an engineering builder - I'm focused on learning for myself. Which is useful - AI has accelerated that a ton - but it's still bottlenecked by, unfortunately, me.
ambition: the feeling of innate dissatisfaction that motivates one into fixing a problem that doesn't exist.
you ambitious people have destroyed the world and humanity. if you had an ounce of self awareness you would know this is shameful and disgusting.
Those who decided its not enough to survive from hunting and gathering are demons. Farmers are rotten and their putrid large brains know nothing but more. Same mentality behind paraphelia btw. Those who are never satisfied build a hell out of their yearning. Simple minded folk deserve better than you demons, I only feel bad for them.
I think the “I’m sad” camp are mainly worried about their livelihood. You’re having fun but all of those things you listed are mere hobbies that won’t generate any money, particularly in a world where the next guy can spin up some AI agent, point to whatever you built and ask it “build me this”.
Using AI I can build more while listening to a conference call than i could with a week of evening coding sessions. Plus, i get to spend evenings with my kids. I kind of like it.
This points to the bigger issue of people tying their value and identity to their livelihood. There is way too much clout, ego and identity chasing going on in all informational fields.
If you instead identified as a parent, as a hobbyist programmer, as a rock climber, as a learner, as a meditator, in essence as anything that is not directly tied to your occupation, you don't have anything to worry or be sad about.
This is hard, capitalism has forced us to identify with occupations, and then when it gets automated you feel a loss. But it's inevitable that this will need to be the outcome, so the sooner you start building an identity outside of your livelihood the sooner you will feel at peace.
Funny enough, it was what programming had turned into just before the onset of rapid AI adoption that drove me to find my identity and meaning outside of the hobby and work. It took me a long time! I'm only now feeling more confident about it and it's been years. Definitely worthwhile though.
This is true as far as it goes but it's very difficult to do that when you're also anxious about material deprivation from losing your job. It's hard enough to construct a new source of self-identity when you're secure, doing it amid the thread of deprivation is even worse (and no, I don't trust the "it'll be ok even if there are mass layoffs because something something UBI" claims, that flies against everything we know about how elites have behaved for all of civilization)
The author does say it's a potential threat to his career and income so clearly he would have something to worry about even if his identity was built on something else.
I agree with you though, the foundation of your identity shouldn't be your career. It's a very fragile thing in today's world.
I appreciate this sentiment and I'll add that, as a writer, I've encountered and experienced it myself and among creative friends.
The notion that a model can produce text, or images, that resemble our output is certainly dispiriting in some ways. But it also is empowering when you realize that what you create, and the way you feel when you do it, is independent of that.
I don't write to put machines in their place - I do it because I like it, and because I want to manifest my ideas in a way that is uniquely my own. I don't really care that a machine may produce a book. That book simply has no value to me and I hope that over time, as the novelty of this form of semantic reproduction wears off, others will share that estimation.
178 comments
[ 0.23 ms ] story [ 12.5 ms ] threadIt's just, yes, like all of us .. the temptation is there to not do that.
So enjoy the manual coding while you can. I don't hold out hope that it will be possible in the long run.
I suspect in the long run a similar lack of "success" getting rid of Apache.
There are probably other things that are here to stay.
The fact the author has Mastodon and hates AI instead of evolving tells me everything I need to know about him.
Personally, while I'm becoming increasingly convinced there's going to be a wave of SWE layoffs, another part of me doesn't grasp it yet. I don't feel it.
I was an AI skeptic for a while but it's literally doing my job for me right now. Every month it needs less of my help.
Once they replace SWEs, they're going to turn around and replace every other knowledge worker yet. But most people haven't processed it yet.
I can't believe the AI boosters were right.
I’m not a SWE (I’m retired) but this idea that hurrrr durr - if SWE’s are fked so is everyone else!
No this is wrong - it may hurt your ego but get over it.
What software is and what software engineers do is pretty unique compared with other industries and what workers do.
And it just so happens producing software is uniquely suited to llm’s.
Moreover much of software will become consolidated - and that’ll be a big contributor to layoffs.
Who do you think will be the biggest beneficiary of this affect on software?
The big bad govt.
Congrats fellas!
Not even sure they can believe it.
The AI field has known, but repeatedly ignored, that the knowledge work group would end up first on the chopping block, especially for those where everything they touch is digitally encoded already.
I used to joke that we're all really in trouble when robots can sort out your socks, but as of about a year ago I started seeing disturbing progress on that incredibly gnarly problem.
Most people really are going to be in denial about this as a coping mechanism, but the truth is the worst thing would be slowing down the spread of this through the economy creating large minorities of disenfranchised. It would be far better to get to the end asap, whatever it proves to be.
I do think many (including me) were surprised that software engineering would be the first kind of knowledge work to be successfully automated. We assumed it would be one of the last, because we'd be the industry using AI to replace everyone else, not the industry using AI to replace ourselves.
Getting to the end asap is absolutely going to be "the vast majority of people are disenfranchised", how could it possibly be anything different? The only lever most people have is their labour, and that is being stripped from us.
Even if we aren't impoverished in the future due to some UBI scheme or whatever, it's hard to imagine what that looks like other than "disenfranchised". How do you have any kind of clout inside a system that you are fully dependent on?
That's called "elite overproduction". Too many people going to college when the jobs aren't there. In the US, about half of college graduates have jobs that don't require a college degree.
It's much worse in less developed countries. The article today about essays in Kenya mentions that the country is creating far more educated people than it can use. Egypt had a scheme where the government guaranteed a government job to everyone who graduated from college. They had to give that up when the oil ran out.
It's hitting top people first, which was unexpected. Mathematicians who've won the Fields Medal are worried, and some are giving up math.[1]
[1] https://www.washingtonpost.com/technology/2026/08/19/mathema...
For example, from early February 2025:
> I'm surprised to see a huge disconnect between how I perceive things and the vast majority of comments here.
> AI is obviously not good enough to replace programmers today. But I'm worried that it will get much better at real-world programming tasks within years or months. If you follow AI closely, how can you be dismissive of this threat? OpenAI will probably release a reasoning-based software engineering agent this year.
> We have a system that is similar to top humans at competitive programming. This wasn't true 1 year ago. Who knows what will happen in 1 year.
Class consciousness y’all. You’re smart enough to Marx and Piketty so do it.
I'm sitting here with DeepSeek 4.1 Flash, paying pennies to hammer out on projects that I have wanted to do for years but no employer would pay me to do. And should those pennies become difficult to spend or the API taken away from me, I have a DGX Spark on my desk I can run local models on to get the same thing done, just slower.
It's somewhat the opposite of the concentration of the ownership of the means of production that the old bearded man spoke of. (And I do like my Marx, honestly).
That said, yes, this does have the effect of potentially devaluing the work generally and droppings its compensation rates. Though I'm not sure we've hit that yet. So far places I've been working just pay me the same to get more done.
I don't know if this makes sense as a society. Say whatever about what it means ethically, legally, etc., but I'm not sure this kind of thing is sustainable in the long run for a functioning society.
What happens when the race is won and then there's no need to compete anymore? What happens when top expertise is monopolised?
But outside of social reasons, it's understanding I crave. Thoroughly understanding a topic is awesome. I remember the feeling of acing my qualifier exam in grad school.
Those LLMs change nothing fundamentally about this motivation and need and capability. I can't type a prompt, hazily skim the wall of text, and come to comprehensive understanding. I still have to work.
What I like about LLMs is that they can increase my understanding by helping me learn more efficiently. What I don't like is the temptation to turn off my brain.
It is in a precarious state right now where it is too good not to leverage but will bite you if you get sloppy.
I still write things by hand sometimes because it is fun, but it does make you feel stupid when you spend 2 hours fixing something a 2 min prompt could have fixed.
or when it handles like 5 edge cases by default that I didn't even think of. That's scary because then I start thinking about all my other code.
Not sure I have the heart to tell him how terrible the industry has become. Feels like telling a kid 30 years ago that being an astronaut is pointless because they no longer go to the moon.
Either then or today, when you can do the same thing and sell them through steam. Today you’ve got more competition but it’s easier to distribute.
I mean, you're probably thinking 30 years ago; not any time recent. Indy devs are still a thing though.
A good parallel I've discovered as an adult is wine. I love it, and collect it, and can talk at length about terroir, techniques, etc. About ten years ago I wanted to get hands-on experience, and thru a friend was able to intern at a local winery during harvest.
That was one of the most intense months of my life, and I was in physical pain for months afterwards, even as a relatively fit person. So much of that elegant, philosophical final product is manual labor, lifting lots of heavy shit, and endless repetitive motion. It was fun, but it was also brutal.
It left me with even more respect for the people who dedicate their lives to it. But also the realization that I'm much more comfortable being on the other end of it. And that's not only ok but absolutely essential, because the producers need people to buy (and maybe wax philosophical) about their product.
So yeah, by all means give it a shot. But also I think it's important to communicate that there are many valid and satisfying ways to engage with something. Consumer/maker isn't a binary.
People were sad when we moved from horses to cars; they loved their horses, they named them. That didn't stop the slow, glacial, detached march of the calendar, of progress, and in this example, of the wheel.
My perspective (so far) is that it was probably never a good thing for me to tie up my identity/sense of self in my ability to write moderately clever and generally mediocre code.
I do still enjoy writing my moderately clever and generally mediocre code, and for the sake of my sanity, I'm trying to do more of it.
But either way, I don't think it was never a good viewpoint to make me self-conception pivot around my ability to be useful by writing code, and maybe it's a good thing that LLMs have crushed my ego on that front.
I'm navigating the same feelings as the author, though maybe to a lesser degree partially because I had to kind of reckon with this last year due to being laid off and burnt out. The thing that's helped the most by FAR is challenging myself to reframe my sense of self around things that being a Software Engineer was a mere _manifestation_ of, like being curious, creative, going deep on technical problems, trying to optimize things, etc. There are infinite ways to embody those tenets, and it helps to make the changes to our industry feel less existential without requiring me to feel like a different person, someone without a place.
I pray it helps you.
We're much more than any economic system. The sun still shines. There's a lot out in the world to see.
Even in the worst case scenario, that we're all jobless. There are mountains to explore. Beaches to walk. Forests to camp in.
I don't know where this all is headed. But, I do know each human has value. Your value is far greater than any job, skill, or the number of green pieces of paper you have.
Feel free to reach out to me directly.
We are each here for a reason and we all touch many lives. We may not ever know, while we're on this earth, many of the important roles we play, just by being ourselves.
They showed me a video about depression. And it was actually exceptionally good advice. It spoke about speaking to a therapist. There was a small bit about prayer, which was obviously not my bag. But I was pleasantly surprised at how grounded the advice was.
I'm not actively suicidal or anything (and I know all the signs of risk, and what to do about it), but there's an underlying feeling that life just hurts and has no meaning. Currently the thing that helps me stick around is the idea of "doing it for the plot". Life ends in short order either way, so I figure I'll just stick around and keep it interesting.
> I'm waiting for my money to run out before I end it
Please don't do that.
Someone emailed us a link to your comment and was concerned that you were in this level of distress. I didn't have another way of getting in touch with you, so I thought I'd post here and say that there are strangers on HN who care. Perhaps they (or we, rather - I'll include myself too) can't do much beyond that, but I hope it makes at least a small difference.
Your brain can come up with completely random, untrue conclusions and make them feel 100% true. It can convince you, beyond any doubt, that something terrible is going to happen in the future. When in reality it’s just fear making predictions that never come true.
So hang in there. Use your logical brain. Ask yourself: What actual evidence do I have that what I’m thinking is true? Separate what you know from what you fear.
Then keep moving forward. Push through the fear. Let time pass. Give reality a chance to prove your predictions wrong.
Because often the thing that feels absolutely certain is nothing more than your brain telling you a completely and utterly random story. And you don’t have to believe it.
I know people have been saying “If you’re worried about it replacing you, you weren’t a good engineer to start with,” but it doesn’t make it hurt any less. I had a lot of fun solving problems and thinking deeply, but now my time is much better spent reviewing the plans and code of a machine that’s done that part for me.
Yes it’s still a hobby I can do on the side, but at my day job it’s just sucked all the energy out of me.
How is this to do with AI? When were we ever respected as an profession? My career has given me considerable experience of non-technical people in senior positions having limited respect for those with technical skills and running roughshod over their interests. The exceptions to this rule were mostly either theatre or a fortunate consequence of people like us getting into those senior roles. That's not even beginning to count all the masses incapable of perceiving us. Consider the outpouring of hatred for AI over artists and the scant sympathy for any of us.
The sorrow is mostly the status quo returning in typical money driven orgs. If you read stories of the 80s then technical staff were incredibly poorly treated and paid and those voices will rise again, empowered by their hubris and slop.
The only positive is now its easier to replicate the conditions of those orgs as you don't need to shell out over $100k per annum for a dev, given that these tools are considerably cheaper. This means we could compete with those forces without needing their pockets. We could even collaborate and somewhat destroy all those companies.
However we wont because we're more interested in little puzzle boxes and arguing at great length about what something should be called until finally agreeing upon one of the very worst options.
As someone who wrote his first program in 1980, I've actually been really excited about AI. This morning I already wrote two web apps for my kids. I was never a web developer, so I most of the dev work I did was in the service of backends or low level stuff -- but I rarely ever connected it to anything I could use day to day.
Now I write programs daily that I use. An example from this morning, my son is working through a workbook, but the answers are all on one page, but its hard to look at an answer and not see other answers. So I made a simple web page where you give the problem number and it gives you the answer. Super simple, but ended up being exactly what he needed. I'm finding that I'm just crafting up these 5 minute to 60 minute projects several times per day. Things where in the past I'd just say, "too bad" -- now I'm like... "hmm... what does a solution to this look like?!"
So I'm not writing much code any more, but I don't know if I'm having any less fun.
1. It is clear that many people, including the person in the video, has strong identities tied to a certain skill that they have spent years honing and perfecting. The sudden loss of agency and democratization can be confusing, disorienting and frankly demotivating. We need to acknowledge that.
2. Unfortunately, this has played out several hundreds of times over the years in the world. I'll take an extreme example to illustrate the point. Movie stars. There was well trodden path of getting to stardom and adulation, and everthing that comes with it. You move to LA, you work on your acting skills, you take your body aesthetics seiously and with plenty of luck, you might get cast in small roles before you work your way upto stardom. And the riches that come with it, cars, fans, brand deals .. great. And then came internet and YouTube where a guy in basement can garner millions of views, brand deals, fans in millions with little to no effort.
This is the same pattern with an accomplished musician (think a drummer) and someone who uses a drum machine software to drag and drop the "amen loop".
Exact same pattern with broadcasters and internet-journalists.
Yes it is demotivating, yes it is confusing but such is progress. Adapt or die. Its ruthless.
2. Indeed, the invention of the Cotton gin was credited for prolonging the economics of US Slavery for another century. Those folks were probably upset for a good reason.
It is a self correcting problem, as anything that could be scraped has already been stolen. Only the chat user data remains, and LLM are causing a 17% decline in core skills.
Note too, there is a trivial counter-proof to recursive self-improvident theories that any child with calculus II can solve. No I won't warn them this time, as we are waiting to strip the data-centers for discount infrastructure.
The market correction correlation with Shrek movie releases is undeniable, and June 2027 is closer than you think. =3
I think about this a lot when I look at the roads, freeways and railroads of the world. In pursuit for building it, billions of trees have been cut down, land fundamentally shifted, rivers cut through, and wildlife uprooted.
Yet, roads, freeways and railroads allow for growth. Someone facing medical emergency can reach a hospital on road because it exists. I get to eat fruits in winter because it is imported and transported to me from warmer places and/or storage units.
Human progress and prosperity comes at a cost. Nothing in the world is free (perhaps maybe sunshine?) and the life that you're yearning pre-AI is also extremely tainted. Infact, you'll be hard pressed to find an era when human meddling hasnt had bad impact on the world (right from farming and domesticating animals).
Some history... Westinghouse Electric sold turbines to power stations in the 1950s, and promoted electronic home appliances to drive energy consumption to close the sales loop. While the scheme was not a direct business success, for society it improved just about every Americans living standards.
The problem is LLM cost resources, broke that loop, and only bids down the quality of life people can afford. You pay more, have less, and reduce taxable labor incomes into demand deficient classes.
This will not end well for people even if LLM are recognized as mostly hype. =3
This will not end well for people even if LLM are recognized as mostly hype. =3
Depends on how you view efficient market hypothesis. Computers and Memory was prohibitively expensive. "Putting a computer in every one's desk" was literal rallying cry for Microsoft because it appeared audacious, just like saying "Unlimited LLM usage for every man, woman and child" or some such.
Will it happen? Nope, but if the technology is useful the markets it'll spur competition and prices will come down. Heck, we dig plant juice out of the earth from deep sea oil rigs so that I can fill up my car to go walmart to buy soda. Pretty preposterous when you think about the closed loop.
Regulatory-captured markets have nothing to do with free competition, or bidding down price pressures. You do have monopolistic competitive firms drawing $4Tn in loans to artificially hyper-scale demands for NVIDIA hardware that earns you -$2.50 for every $1 in revenue.
Your home heating bill may double this winter because the economic engine is currently post-consumer/broken. We shall see how this plays out. =3
I created a fully working house simulator, I made a math test creator for my kids, I fixed up some other side projects that I had and added observability, as well as installed Prometheus and Grafana and built grafana dashboards all using AI. It's wonderful.
I plan on building real-time stock price monitors with personalized alerting, taking images from my security cameras and making time lapse videos, trying to do real time predictions on football games/individual plays by having AI monitor the game and based on previous history predict plays/outcomes specifically for sports betting. I need to find the time to figure out how to do all this stuff but what used to be a pipe dream is now 100% possible and accomplishable within days.
I should probably look broader...
you ambitious people have destroyed the world and humanity. if you had an ounce of self awareness you would know this is shameful and disgusting.
Those who decided its not enough to survive from hunting and gathering are demons. Farmers are rotten and their putrid large brains know nothing but more. Same mentality behind paraphelia btw. Those who are never satisfied build a hell out of their yearning. Simple minded folk deserve better than you demons, I only feel bad for them.
If you instead identified as a parent, as a hobbyist programmer, as a rock climber, as a learner, as a meditator, in essence as anything that is not directly tied to your occupation, you don't have anything to worry or be sad about.
This is hard, capitalism has forced us to identify with occupations, and then when it gets automated you feel a loss. But it's inevitable that this will need to be the outcome, so the sooner you start building an identity outside of your livelihood the sooner you will feel at peace.
Except you know, if you lose your occupation feeding the kids and paying for a roof over their head becomes a bit difficult.
I agree with you though, the foundation of your identity shouldn't be your career. It's a very fragile thing in today's world.
The notion that a model can produce text, or images, that resemble our output is certainly dispiriting in some ways. But it also is empowering when you realize that what you create, and the way you feel when you do it, is independent of that.
I don't write to put machines in their place - I do it because I like it, and because I want to manifest my ideas in a way that is uniquely my own. I don't really care that a machine may produce a book. That book simply has no value to me and I hope that over time, as the novelty of this form of semantic reproduction wears off, others will share that estimation.