Why LLMs can't really build software

Interesting what he wrote about LLMs' inability to "zoom out" and see the whole picture. I use Gemini and ChatGPT sometimes to help debug admin / DevOps problems. It's a great help for extra input, a bit like rubberducking on steroids.

Examples how it went:

Problem: Apache-cluster and connected KeyCloak-Cluster, odd problems with loginflow. Reducing KeyCloak to 1 node solves it, so it says that we need to debug node communication and how to set the debug log settings. A lot of analysis together. But after a while, it's pretty obvious that the Apache-cluster doesn't use the sticky session correctly and forwards requests to the wrong KeyCloak node in the middle of the login flow. LLM does not see that, wanted to continue to dig deeper and deeper into supposedly "odd" details of the communication between KeyCloak nodes, althought the combined logs of all nodes show that the error was in load balancing.

Problem: Apache from a different cluster often returns 413 (payload too large). Indeed it happens with pretty large requests, the limit where it happens is a big over 8kB without the body. But the incoming request is valid. So I ask both Gemini and ChatGPT for a complete list of things that cause Apache to do that. It does a decent job at that. And one of it is close: It says to check for mod_proxy_ajp use, since that observed limit could be caused by trying to make an AJP package to communicate with backchannel servers. It was not the cause; the actual mod was mod_jk, which also uses AJP. It helped me focus on watching out for anything using AJP when reviewing the whole config manually, so I found it, and the "rubberducking" helped indirectly. But the LLM said we must forget about AJP and focus on other possible causes - a dead end. When I told it the solution, it was like: Of course mod_jk. (413 sounds like the request TO the apache is wrong, but actually, it tries internally to create an invalid AJP package over 8kB, and when it fails blames the incoming request.)

LLMs are useful to provide generic examples of how a function works. This is something that would previously take an hour of searching the docs and online forums, but the LLM can do for very quickly, and I appreciate. But I have a library I want to use that was just updated with entirely new syntax. The LLMs are pretty much useless for it. Back to the docs I go! Maybe my terrible code will help to train the model. And in my field (marine biogeochemistry), the LLM generally cannot understand the nuances of what I'm trying to do. Vibe coding is impossible. And I doubt the training set will ever be large or relevant enough for the vibe coding to be feasible.

Thats simply not true. LLMs with RAG can easily catch up with new library changes.
- Subjectively speaking, I don't see it so that good a job of being current or priortizing current over older.
  While RAG is the way to give LLM a shot at staying current, I just didn't see it doing that good a job with library documentation. Maybe it can do all right with tweaks like additional properties or arguments, but more structural changes to libraries I just don't see being handled.
- Yes and no.
  In many cases (like for the Gradle DSL, that even if it can be either the old Groovy-based one or the new Kotlin-based one, you will always be able to find extensive documentation and examples in the wild for both of them) it is sufficient to specify which version you're using and, as long as this doesn't get too far in its context window forcing you to repeat it, you are good to go.
  But for niche libraries that have recently undergone significant refactors with the majority of the tutorials and examples still built with past versions, they have a huge bias towards the old syntax, making it really difficult - if not impossible - to make them use the new functions (at least for ChatGPT and GitHub Copilot with the "Web search" functionality on).
You can't know without checking though, it may be wrong
Vibe coding
The term for that is actually 'slopping'. Kthx ;-)

Note: this comes from someone that makes a (very good) ide which they only monetize with an AI subscription so it's interesting to see their take

(They use Claude opus like all the others so the results are similar)

I think AI in you IDE is meant to help you with small things while AI agents are supposed to do development for you. If people will start using AI agents they won't need IDEs so this take is consistent with their business model.
in one regard I can understand, they're running a business and don't want to be at a disadvantage against their competition.
on the other hand have some conviction for your product, otherwise I will lose confidence that your product is as good as your marketing makes it seem.
- They are still bullish on LLM, just to augment rather than displace human suggested development.
  This perspective is quite consistent with the need for a product that manages prompting/context for a human user and helps the human review and integrate the LLM supplied content in a reasonable way.
  If LLM were as useful as some of the fanatics say, you'd just use a generic prompt and it would poop out the finished project. This is by the way the perspective of an executive I talked to not long ago, that he was going to be able to let go of all his "coders" and feed his "insight" directly into a prompt that will do it all for him instead. He is also easily influenced so articles like this can reshape him into a more tenable position, after which he'll pretend he never thought a generic prompt would be good enough

LLMs have made it really clear when previous concepts actually grouped things that were distinct. Not so long ago, Chess was thought to be uniquely human, until it wasn't, and language was thought to imply intelligence behind it, until it wasn't.

So let's separate out some concerns and ask what exactly we mean by engineering. To me, engineering means solving a problem. For someone, for myself, for theory, whatever. Why do we want to solve the problem, what we want to do to solve the problem, and how we do that often blurred together. Now, AI can supply the how in abundance. Too much abundance, even. So humans should move up the stack, focus on what problem to solve and why we want to solve it. Then, go into detail to describe what that solution looks like. So for example, making a UI in Figma or writing a few sentences on how a user would actually do the thing. Then, hand that off to the AI once you think it's sufficiently defined.

The author misses a step in the engineering loop that's important though. Plans almost always involve hidden assumptions and undefined or underdefined behavior that implementation will uncover. Even more so with AI, you can't just throw a plan and expect good results, the humans need to come back, figure out what was underdefined or not actually what they wanted, and update the plan. People can 'imagine' rotating an apple in their head, but most of them will fail utterly if asked to draw it; they're holding the idea of rotating an apple, not actually rotating the apple, and application forces realization of the difference.

The author misses a step in the engineering loop that's important though. Plans almost always involve hidden assumptions and undefined or underdefined behavior that implementation will uncover.
His whole point is two mental models and a model delta. Exactly what you just described.

I've done a test of 8 LLMs, on coding. It was using the J language, asking all of them to generate a chess "mate in x solver"

Even the bad models were good at organizing code, and had some understanding of chess, were good at understanding the ideas in their prompts. The bad models were bad mostly on logic. Not understanding indexing/amend on a table, not understanding proper function calling, or proper decomposition of arguments in J. Bad models included copilot and openAI's 120g open source model. kimi k2 was ok. Sonet 4 the best. I've mostly used Qwen 3 245 for better free accessibility than Sonet 4, and the fact that it has a giant context that makes it think harder (slower) and better the more its used on a problem. Qwen 3 did a good job in writing a fairly lengthy chess position scoring function, and then separating it into 2 quick and medium function, incorporating self written library code, and recommending enhancements.

There is a lot to get used to in working with LLMs, but the right ones, can generally help with code writting process. ie. there exists some code outputs which even when wrong, provide a faster path to objectives than if that code output did not exist. No matter how bad the code outputs, you are almost never dumber for having received it, unless perhaps you don't understand the language well enough to know its bad.

I don't work in IT, but I do know you need creativity to work in the industry, something which the current LLM/AI doesn't possess.

Linguists also dismiss LLMs in similar vein because LLMs can't grasp context. It is always funny to be sarcastic and ironic on an LLM.

Soft skills and culture are what that the current iteration of LLMs lack. However, I do think there is still huge potential for AI development in dacades to come, but I want this AI bubble to burst as "in your face" to companies.

The LLM worship has to stop.

It's like saying a hammer can build a house. No, it can't.

It's useful to pound in nails and automate a lot of repetitive and boring tasks but it's not going to build the house for you - architect it, plan it, validate it.

It's similar to the whole 3D printing hype. You can 3D print a house! No you can't.

You can 3D print a wall, maybe a window.

Then have a skilled Craftsman put it all together for you, ensure fit and finish and essentially build the final product.

You’re making a great analogy with the 3D printing of a house.
However, if we consider the 3D printed house scenario; that skilled craftsman is now able to do things on his own that he would have needed a team for in the past. Most, if not all, of the less skilled members of that team are not getting any experience within the craft at that point. They’re no longer necessary when one skilled person can now do things on their own.
What happens when the skilled and highly experienced craftsmen that use AI as a supplemental tool (and subsequently earn all the work) eventually retire, and there’s been no juniors or mid-levels for a while? No one is really going to be qualified without having had exposure to the trade for several years.
- Absolutely. This is a huge problem and I've read about this very problem from a number of sources. This will have a huge impact on engineering and information work.
  Interestingly enough, A similar shortage occurred in the trades when information work was up and coming and the trades were shunned as a career path for many. Now we don't have enough plumbers and electricians. Trades are now finding their the skills in high demand and charging very high rates.
I hate the simulated intelligence nonsense at least as much as you, but you should probably know about this if you’re saying you can’t 3d print a house: https://youtu.be/vL2KoMNzGTo
- Yeah I've seen that before and it's basically what I'm talking about. Again, that's not "printing a 3D house" as hype would lead one to believe. Is it extruding cement to build the walls around very carefully placed framing and heavily managed and coordinated by people and finished with plumbing, electrical, etc.
  It's cool that they can bring this huge piece of equipment to extrude cement to form some kind of wall. It's a neat proof of concept. I personally wouldn't want to live in a house that looked anything like or was constructed that way. Would you?
3d printed concrete houses exist. Why can't you 3d print a house? Not the best metaphor lol
- You can certainly 3D print a building, but can you really 3D print a house? Can it 3d print doors and windows that can open and close and be locked? Can it 3D print the plumbing and wiring and have it be safe and functional? Can it 3D print the foundation? What about bathroom fixtures, kitchen cabinets, and things like carpet?
  It's actually not a bad metaphor. You can use a 3D printer to help with building a house, and to 3D print some of fixtures and bits and pieces that go into the house. Using a 3D printer would automate a fair amount of the manual labor that goes into building a house today (at least how it is done in the US). But you're still going to need people who know what they are doing put it all together to transform the building to a functional home. We're still a fair ways away from just being able to 3D print a house, just like we're fair ways away from having a LLM write a large, complex piece of software.
- No they aren't. With enough setup and very unique and expensive equipment, you can pour shitty concrete walls that will be way more expensive and worse than if you did it normally. That will give you 20% of the house, at best. 20% of not very good of a house.
- You don't like glass windows? Air conditioning? A door?

I think it's going to require a change in how models are built and optimized. Software engineering requires models that can do more than just generate code.

You mean to tell me that language models aren't intelligent? But that would mean all these people cramming LLMs in places where intelligence is needed are wasting their time?? Who knew?

Me.

I have a solution for that, I just need a small loan of a billion dollars and 5 years. #trustmebro
- Only one billion?? What a deal! Where's my checkbook!?

Clearly LLMs are useful to software engineers.

Citation needed. I don’t use one. If my coworkers do, they’re very quiet about it. More than half the posts I see promoting them, even as “just a tool,” are from people with obvious conflicts of interest. What’s “clear” to me is that the Overton window has been dragged kicking and screaming to the extreme end of the scale by five years of constant press releases masquerading as news and billions of dollars of market speculation.

I’m not going to delegate the easiest part of my job to something that’s undeniably worse at it. I’m not going to pass up opportunities to understand a system better in hopes of getting 30-minute tasks done in 10. And I’m definitely not going to pay for the privilege.

I’m not a “software engineer” but a lot of people that don’t work within tech would probably call me one.
I’m in Cloud Engineering, but came from the sys/network admin and ops side of things rather than starting off in dev or anything like that.
Up until about 5 years ago, I really only knew Powershell and a little bit of bash. I’ve gotten up to speed in a lot of things but never officially learned python, js, go or any other real development language that would be useful to me. I’ve spent way more time focusing on getting good with IaC, and probably more of the SRE type stuff.
In my particular situation, LLMs are incredibly useful. It’s fair to say that I use them daily now. I’ve had it convert bash scripts to python for me very quickly. I don’t know python but now that I’m able to look at a python script next to my bash; I’m picking up on stuff a lot faster. I’m using Lambda way more often as a result.
Also, there’s a lot of mundane filling out forms shit that I delegate to an LLM. I don’t want to spend my time filling out a form that I know no one is actually going to read. F it, I’ll have the AI write a report for an AI. It’s dumb as shit, but that’s the world today.
I've only found two effective uses for them. Every time I tried them otherwise they fell flat and took me longer that it would have to write the code myself.
The first was a greenfield personal project where I let code quality wane since I was the only person maintaining it, and wanted to test LLMs. The other was to write highly repeative data tests where the model can simply type faster than me.
Anything that requires writing code that needs to be maintained by multiple people or systems older than 2 years, it has fallen completely flat. In cases like that I spend more time telling the LLM it is doing it wrong, it would have taken me less time to write the code in the first place. In 95% of cases, I am still faster than an LLM at solving a problem and writing the code.
I've found them useful, sometimes, but nothing like a fraction of what the hype would suggest.
They're not adequate replacements for code reviewers, but getting an AI code review does let me occasionally fix a couple of blunders before I waste another human's time with them.
I've also had the occasional bit of luck with "why am I getting this error" questions, where it saved me 10 minutes of digging through the code myself.
"Create some test data and a smoke test for this feature" is another good timesaver for what would normally be very tedious drudge work.
What I have given up on is "implement a feature that does X" questions, because it invariably creates more work than it saves. Companies selling "type in your app idea and it'll write the code" solutions are snake-oil salesman.
I don't use one, and my coworkers that do use them are very loud about it, and worse at their jobs than they were a year ago.
I have been using it a bit, still can't decide if it is useful or not though... It can occasionally suggest a blatantly obvious couple of lines of code here and there, but along the way I get inundated with annoying suggestions that are useless and I haven't gotten used to ignoring them.
I mostly work with a niche area the LLMs seem broadly clueless about, and prompt driven code is almost always useless except when dealing with a super boilerplate usage of a common library.
I do know some people that deal with amazingly mundane and common functions and they are amazed that it can pretty much do their jobs, but they never really impressed me before anyway and I wondered how they had a job...
https://survey.stackoverflow.co/2025/ai/
47% daily use
- 47% daily use
  That is NOT what that says. It says 47% of STACK OVERFLOW RESPONDENTS REPORT using AI. That does not represent 47% of devs.
  If you go to 4chan and poll of chuds, you're going to get a high percentage of respondents affirming your query. You went to stackoverflow and asked about AI. Think about the user base.
If my coworkers do, they’re very quiet about it.
Gee, guess why. Given the current culture of hate and ostracism I would never outright say IRL that I like it or use it a lot. I would say something like "yeah, I think it can sometimes be useful when used carefully and I sometimes use it too". While in reality it would mean that it actually writes 95% of code under my micromanagement.
- Wut. At software shops the prevailing atmosphere is that you should use it and broadcast it as much as possible. This person's experience is not normal

To those who have played around with LLM code generation more than me, how are they at debugging?

I'm thinking of Kernighan's Law: "Debugging is twice as hard as writing the code in the first place. Therefore, if you write the code as cleverly as possible, you are, by definition, not smart enough to debug it." If vibe coding reduces the complexity of writing code by 10x, but debugging remains just as difficult as before, then Kernighan's Law needs to be updated to say debugging is 20x as hard as vibe coding. Vibe coders have no hope of bridging that gap.

I have next to zero experience about coding (unless you count a few months of Borland Delphi work back in 00’s, which you shouldn’t). Yet, I’ve managed to create half a dozen really useful tools for my work, and a few more for my hobbies too.
Inflection point for me was Gemini 2.5 Pro. Before that I was only successful with smaller scripts, using ChatGPT mostly. But with Gemini I was able to do Deep Research as the initial step to plan out the overall architecture, interfaces, technologies etc. and finetune the actual coding prompt using that info.
Crucial step after first generated (buggy) version is to copy paste the code and errors to ChatGPT and/or Grok to get their take on it, then feed back those ideas to Gemini again. Some 5-10 iterations of this and I usually have a fully functional application or a component of bigger piece software. Problems at the moment usually arise if any particular file exceeds ~800 lines, and when there are many many iterations. Then LLMs tend to get forgetful, dropping out comments, reintroducing faults from earlier iterations etc. Better to start a new session at that point.
Thinking of LLMs as just a lossy compression algo for all human knowledge, the parallel use of LLMs makes kind of sense: All the companies use approximately the same data in their training, but end up having a bit different looking ”lossy big picture” in the end. But if I ”look at all these pictures” side by side I can perhaps see more detail. Or that some of the pictures are fuzzy on one location but one is much clearer.
LLM’s seem to be very good to at spotting the correct solution when they are given a couple options or hypothesis about an cause of an issue. Most surprising to me is that Grok has been the one to solve majority of the most stubborn bugs that other have gotten stuck to.
With GPT 5 I just had my first ”hole in one” where it generated a flawless ~500 line web app in the very first try. Just gave it my Git codebase as a zip file and asked for a new module that interfaces the existing stuff. Wild times.
I've used AI by just pasting code, then asking if there's anything wrong with it. It would find things wrong with it, but would also say some things were wrong when it was actually fine.
I've used it in an agentic-AI (Cursor), and it's not good at debugging any slightly-complex code. It would often get "stuck" on errors that were obvious to me, but making wrong, sometimes nonsensical changes.
I am working at a big AI company on llm generating code for automation. I've had cursor solve a bug that was occuring in prod after prompting and asking questions of the responses. It took a few rounds but it found a really obscure interaction with the app and the host, and it thanked me for the insight. 😀. I deployed the fix and it worked.
The problem I have is I member it solving this bug, and I remember being impressed, but I don't remember the bug. I took a screenshot of it, but currently don't have access to those. I am disconnected from the code that the llm has generated, but I am very aware of how the app works and what it should do very intently because I had to write requirements and design doc.
How are they at debugging? In a silo, they’re shit.
I’ve been using one LLM to debug the other this past week for a personal project, and it can be a bit tedious sometimes, but it eventually does a decent enough job. I’m pretty much vibe coding things that are a bit out of my immediate knowledge and skill set, but I know how they’re supposed to work. For example, I’ve got some python scripts using rekognition to scan photos for porn or other explicit stuff before they get sent to an s3 bucket. After that happens, there’s now a dashboard that’s going to give me results on how many images were scanned and then marked as either acceptable or flagged as inappropriate. After a threshold of too many inappropriate images being sent in, it’ll shadowban them from sending any more dick pics in.
For someone that’s never taken a coding course, I’m relatively happy with the results I’m getting so far. Granted, this may be small potatoes for someone with an actual development background; but as someone that’s been working adjacent to those folks for several years, I’m happy with the output.
They're not good at debugging. The article is pretty spot on, IMO - they're great at doing the work; but you are still the brain. You're still deciding what to do, and maybe 50% of the time how to do it, you're just not executing the lowest level anymore. Similar for debugging - this is not an exercise at the lowest level, and needs you to run it.
- deciding what to do, and maybe 50% of the time how to do it, you’re just not executing the lowest level anymore
  And that's exactly what I want. And I don't get it why people want more. Having more means you have less and less control or influence on the result. What I want is that in other fields it becomes like it is in programming now, so that you micromanage every step and have great control over the result.
The company I work for has recently mandated that we must start using AI tools in our workflow and is tracking our usage, so I've been experimenting with it a lot lately.
In my experience, it's worse than useless when it comes to debugging code. The class of errors that it can solve is generally simple stuff like typos and syntax errors — the sort of thing that a human would solve in 30 seconds by looking at a stack trace. The much more important class of problem, errors in the business logic, it really really sucks at solving.
For those problems, it very confidently identifies the wrong answer about 95% of the time. And if you're a dev who's desperate enough to ask AI for help debugging something, you probably don't know what's wrong either, so it won't be immediately clear if the AI just gave you garbage or if its suggestion has any real merit. So you go check and manually confirm that the LLM is full of shit which costs you time... then you go back to the LLM with more context and ask it to try again. It's second suggestion will sound even more confident than the first, ("Aha! I see the real cause of the issue now!") but it will still be nonsense. You go waste more time to rule out the second suggestion, then go back to the AI to scold it for being wrong again.
Rinse and repeat this cycle enough times until your manager is happy you've hit the desired usage metrics, then go open your debugging tool of choice and do the actual work.
- maybe its just me but I find typos to be the most difficult because my brain and easily see it as correct so the whole code looks correct. Its like the way you can take the vowels out of sentences and people can still immediately read it.
- we must start using AI tools in our workflow and is tracking our usage
  Reads to me as "Please help us justify the very expensive license we just purchased and all the talented engineers we just laid off."
  I know the pain. Leadership's desperation is so thick you can smell it. They got FOMO'd, now they're humiliated, so they start lashing out.
- As it seems to be the case in all of these situations, AI fails hard at tasks when compared to tools specifically designed for that task. I use Ruff in all my Python projects because it formats my code and finds (and often fixes) the kind of low complexity/high probability problems that are likely to pop up as a result of human imperfection. It does it with great accuracy, incredible speed, using very little computing resources, and provides levels of safety in automating fixes. I can run it as an automation step when someone proposes code changes, adding all of 3 or 4 seconds to the runtime. I can run it on my local machine to instantly resolve my ID10T errors. If AI can't solve these problems as quickly, and if it can't solve anything more complicated reliably, I don't understand why it would be a tool I would use.
Definitely not good. Sometimes they can solve issues but you gotta point them in the direction of the issue. Other times they write hacky workarounds that do the job for the moment but crash catastrophically with the next major dependency update.
- I saw an LLM override the casting operator in C#. An evangelist would say "genius! what a novel solution!" I said "nobody at this company is going to know what this code is doing 6 months from now."
  It didn't even solve our problem.
My first level of debugging is logging things to console. LLMs here do a decent job at "reading your mind" and autocompleting "pri" into something like "println!("i = {}, x = {}, y = {}", i, x, y);" with very good context awareness of what and how exactly it makes most sense to debug print in the current location in code.
I use it extensively daily.
It cannot step through code right now, so true debugging is not something you use it for. Most of the time the llm will take the junior engineer approach of "guess and check" unless you explicitly give it better guidance.
My process is generally to start with unit tests and type definitions, then a large multipage prompt for every segment of the app the llm will be tasked with. Then I'll make a snapshot of the code, give the tool access to the markdown prompt, and validate its work. When there are failures and the project has extensive unit tests it generally follows the same pattern of "I see that this failure should be added to the unit tests" which it does and then re-executes them during iterative development.
If tests are not available or if it is not something directly accessible to the tool then it will generally rely on logs either directly generated or provided by the user.
My role these days is to provide long well thought out prompts, verify the integrity of the code after every commit, and generally just kind of treat the llm as a reckless junior dev. Sometimes junior devs can surprise you, like yesterday I was very surprised by a one shot result: asking for a mobile rn app for taking my rambling voice recordings and summarize them into prompts, it was immediately remarkably successful and now I've been walking around mic'd up to generate prompts.
Working just fine. It one shot a kodi tv channel addon for me last week end. Used it to integrate kofax into docusign. Building 2 blazor apps one new one an upgrade. Used it to create a stack of mc servers for the kids with a dashboard of statuses and control switches. My son is working on his own mc mod with it. Use it almost daily for random file organization and management scripts. Using it to clean uo my media library meta data. Anytime i have to do something to more than 5 or so files i pull it up and ask for a script.
Its a tool like any other. There will be people who adapt and people who fail to. Just like we had with computers the internet. It zeems to be long forgotten now but literally ALL of these anti ai arguments were made against computers and the internet 30_50 years ago. Very similar ones were made when books and writing became common place as well.
- "Some random people were wrong about something in the past so nobody is allowed to speculate that any technology isn't as revolutionary as it's hyped to be ever again" is not a useful or compelling argument.
- Apparently I'm not up to date. I've been impressed by some things and turned off by others. But I haven't seen any workflows or setups that enabled access to my file system. How is that accomplished, and are there any safeguards around it?

Good article, I couldn't agree with it more, it's exactly my experience.

The tech is being developed really fast and that is the main issue when taking about ai. Most ai haters are using the issues we might have today to discredit the while technology which makes no sense to me.

And this issue the article talks about is apparent and whoever solves it will be rich.

However, it's interesting to think about the issues that come next.

Like the guy whose baby doubled in weight in 3 months and thus he extrapolated that by the age of 10 the child would weigh many tons, you're assuming that this technology has a linear rate of improvement of "intelligence".
This is not at all what's happening - the evolution of things like LLMs in the last year or so (say between GPT4 and GPT5) is far less than it was earlier in that Tech and we keep seeing more and more news on problems about training it further and getting it improved, including the big one which is that training LLMs on the output of LLMs makes them worse, and the more the output of LLMs is out there, the harder it gets to train new iteractions with clean data.
(And, interestingly, no Tech has ever had a rate of improvement that didn't eventually tailed of, so it's a peculiar expectation to have for a specific Tech that it will keep on steadily improving)
With this specific path taken in implementing AI, the question is not "when will it get there" but rather "can it get there or is it a technological dead-end", and at least for things like LLMs the answer increasingly seems to be that it is a technological dead-end for the purpose of creating reasoning intelligence and doing work that requires it.
(For all your preemptive defense by implying that critics are "ai haters", no hate is required to do this analysis, just analytical ability and skepticism, untainted by fanboyism)
- The difference here is that the current ai tech advancements are not just a consequence of one single tech, but of many.
  Everything you wrote you believe, depends on this being one tech, one dead end.
  The real situation is that we finally have the hardware and the software to make breakthroughs. There is no dead end to this. It's just a series of steps, each contributing by itself and by learning from its mass implementations. It's like we got he first taste of ai and we can't get enough. Even if it takes a while to the next advancement.
It's true, the tech will get better in the future, we just need to believe and trust the plan.
Same thing with crypto and NFT's. They were 99% scam by volume, but who wouldn't love moving their life savings into a digital ecosystem controlled by a handful of rich gambling addicts with no consumer protections? Imagine, you'll never need to handle dirty paper money ever again, we'll just put it all in a digital wallet somewhere controlled by someone else coughmastercardcough.
And another thing, we were too harsh on the Metaverse. Sure, spending 8 hours in VR could make you vomit, and the avatars made ET for the Atari look like Uncharted 4, but it was just in its infancy!
I too want to outsource all my critical thinking to a chatbot controlled by an wealthy insular narcissist who throws Nazi salutes. The technology just needs time to mature. Who knows, maybe it can automate the exile of birthright citizens for us too!
/s
- That's exactly the hyperbole I was talking about. Your post is full of obvious fallacies, but the fact that you are pushing everything to the absolutes is the silliest one.

I love it how article baits AI-haters to upvote it, even though it's very clearly pro-AI:

At Zed we believe in a world where people and agents can collaborate together to build software. But, we firmly believe that (at least for now) you are in the drivers seat, and the LLM is just another tool to reach for.

How is that pro-AI? It clearly very neutrally says it’s just a tool, which you can also hate.

The idea of the mental model CAN be done by AI.

In my experience, if you get it to build a requirements doc first, then ask it to implement that while updating it as required (effectively it's mental state). you will get a pretty good output with decent 'debugging' ability.

This even works ok with the older 'dumber' models.

That only works when you have a comprehensive set of requirements available though. It works when you want to add a new screen/process (mostly) but good luck updating an existing one! (I haven't tried getting it to convert existing code to a requirements doc - anyone tried that?)

I tried feeding ChatGPT a Terraform codebase once and asked it to produce an architecture diagram of what the code base would deploy to AWS.
It got most of the little blocks right for the services that would get touched. But the layout and traffic direction flow between services was nonsensical.
Truth be told it did do a better job than I thought it would initially.
- The trick is to split up the tasks into chunks.
  Ask it to identify the blocks.
  Then ask it to identify the connections.
  Then ask it to produce the diagram.