Both of them can be true at the same time? Many people on HN are at the forefront of this technology, we're testing it in prod and telling each other what does or doesn't work. It's not anti-AI to use the AI and then document a failure.

We're still waiting for a model that can draw a pelican on a bike, you're not zero-shotting every problem with AI yet. If we want improvement, we gotta start by being honest.

I think it's because too many people have released tools that's clearly not ready for production because they don't know what to actually check. So it's now just easier to pattern match away any good tools that might surface.

Claude calls it enterprise and production ready. I now have to spend the next two days dealing with the fallout, page, outage.

> They care that the product works.

And that's the problem.

You are training your replacement.

Because AI use correlates with sloppiness, and due to the fundamental attribution fallacy us engineers don't like sloppiness.

There are two different crowds using "AI":

- One crowd is using to research algorithms, libraries, write boilerplate code, write test harnesses, introspect and integrate with APIs, do hands-off refactoring, and automating what would otherwise be boring tasks. They still think about architecture, best practices, understanding things in detail and the general shape of the solution is in their hands.

- Another crowd is curating prompts, setting up autonomous agents, creating tooling and guardrails around it, anything else but getting actually involved in how the sausage is made. They are working on meta tasks around the problem, in the hope the solution will write itself.

These two crowds are currently living in very different worlds, and getting very different results. We'll see what survives soon.

AI is great for prototyping, but that is far different to AI in production-grade software, including with the hidden cost of maintenance. You have to know what you are doing.

Why even risk using AI directly in mission critical high risk software powering cars, planes and financial transactions or control systems with no human oversight?

If a disaster happened and an investigation was launched and the inquiry found that the software was "vibe coded" and no-one understood the code, would that look great towards the software vendor's reputation?

I don't think it is a large number of people creating this perception, I think it is more their depth of feeling about the issue.

I am often struck by the similarity with bigotry about migrants, where they are portrayed as unreliable and undtustworthy entities that are threatening jobs. Simultaneously arguing their inability and ability are problematic.

You have a second vein of behaviour that object on more religious grounds. There are people that believe that any real understanding of models would deny biblical truth, much like evolution, it is a spurious claim, but at the same time the Discovery Institute is putting money into AI disinformation.

I am unsure how much the Future of Life Institute has influenced thinking, they reputedly have a war chest of half a billion. I have certainly seen videos on YouTube that have been sponsored by them.

It's simply divided. With every such division A vs. B, the A team thinks HN is anti-A and the B team thinks it's anti-B. This is an invariant.

You can see from this megathread, currently on the front page, that HN is by no means anti-AI:

Ask HN: What was your "oh shit" moment with GenAI? - https://news.ycombinator.com/item?id=48406174.

Sometimes it just takes the right initial condition (e.g. title) to bring out one side or other.

As for why the community is divided, there's always a temptation to come up with HN-specific explanations, but society as a whole is divided about AI. Surely that is the only explanation one needs. As I've been saying for years, we can't be immune from macro trends: https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...

It's way more than code. Sure AI can crank out code at prodigious rates. Gary Tan, Y Combinator's CEO says he ships 37,000 lines of AI code per day [0]

And so can I. (oops)

"In the Beginning" (I was there) people wrote accounting packages in BASIC. It worked, the language allowed rapid prototyping, and out the door quickly, but BASIC lent itself to spaghetti code, and for anything really serious, the programs were too lightweight, and were very difficult to document and maintain, so that bugs could be fixed and esoteric features added (for $$$) without the fix breaking something else. Every damn line of code had to be commented so that someone else could pick it up when you left and maintain and upgrade it.

So, AI's got a mind of its own, and from what I hear, every time you get a solution (code) it's different from the previous. At this point, no solid libraries, such as mathematicians, physicists, medical researchers and yes, rocket scientists can rely on as 100% solid and "bet your life on it"

In addition, the hype has extended AI into more general areas, including "bet your life on it" situations where people are using it for therapy, with fatal consequences at times [1] "Nearly 1 in 5 U.S. Adolescents and Young Adults Use AI Chatbots for Mental Health Advice" (RAND) and it's so flawed.

And also, it leads to cognitive surrender. [2] "AI and the Psychology of Cognitive Surrender" (Psychology Today)

Key points:

  • AI subtly erodes our cognitive strength by making delegation seem like self-generated thought.

  • After repeatedly turning to AI for answers, the first thing that erodes is tolerance for not knowing.

  • True judgment is built by wrestling with uncertainty, not outsourcing discomfort to machines. 

In a very brief thread about Siri becoming AltSiri [3] my comments regarding the wide use of a tool that is IMO overextended and using the general population as guinea pigs:

---

I view and use computers as tools. They (mostly) do what I command.

That's because I am by nature a problem solver, and so are others. In fact, if knowledge consists of understanding a particular domain, and wisdom consists of applying knowledge across different domains, creativity of a sort, one of them being that unknown called the future then "button pusher" answers kill my ability to deal with future situations which are not recorded in "The Book of Common Knowledge" (a SNL reference).

When "computers" wrestle control of the situation and solve everything, then, as someone said in the early 20th century "Everything that can be invented has already been invented" then there's now no need for computers at all, since "Every problem can be solved by a chatbot" and no need for creative (genius) things like the famous "Wordless Workshop" that ran in Popular Science and Family Handyman magazines.

Just answer machines. No need to learn anything, nor to create.

Creativity and genius move us forward. That's why we have Hacker News as opposed to those "answer forums"

---

And YES, code that you have to reverse engineer in order to maintain must be understandable and well-architected. If that's "Elegant" then So be it.

I rapidly prototyped in-house apps, quickly and well, and they had a limited life span.

But "enterprise" software isn't going away. And whom [4] do you call when some CTO calls you at 1 a.m. when their business takes a header? Claude?

[0] https://news.ycombinator.com/item?id=48414607

[1] https://www.rand.org/news/press/2026/06/nearly-1-in-5-us-ado...

[2] https://www.psychologytoday.com/us/blog/the-digital-self/202...

[3] https://news.ycombinator.com/item?id=48413555

[4] I was born in Boston. Cheers!

The reports of AI powered 10x development speed are greatly exaggerated

HN is not anti-AI. HN reflects a reasonable ratio of pro-AI and anti-AI sentiments (sometimes held by the same person! because AI covers a lot of ground).

Most people in general have a negative view on AI, the HN crowd isn’t special

Mainly because noisy people are most visible. Both pro-AI and anti-AI (so to speak) crowds have them.

AFAICT hacker news is only slightly less positive on AI than the average tech industry gathering, which is still like two standard deviations more positive than any average gathering of random people in a city. I think the culture of silicon valley reads anything less than gushing hype as negativity right now, which is a weirdly polarized place to be, but the discourse around this technology is bizarre in general, being an absolute gamechanger that nonetheless still somehow feels quite oversold by its most ardent boosters, who are themselves a minority, but one with rather disproportionate voice and reach

You have no obligation to agree with them, but after all this time I don't know how someone on either side could be ignorant of what the other side's main arguments are.

A lot of people on HN are anti-overhyping, which comes across as being opposed to the thing being overhyped. It was similar when cryptocurrency overhyping was popular.

The root cause if of course AI's role in loss of power on compensation (coding as a skill is no longer as valuable), and loss of power in labor vs capital.

It's hard to face this, specially for the one oasis in the job market that pays well.

> They care that the product works

This reminds me of Anthropic's post where they say they ship 8x as much code as they used to.

And I stopped to consider how many times I've used an app and thought, "You know what this needs? More code!"

wth are you talking about

Isn't the mere fact that every HN frontpage is filled with AI-related articles not indicative enough of how much it holds interest here?

> post about how AI “writes bad code,” “introduces bugs,” “creates technical debt,” or something along those lines.

Many people here are engineers and are interested in solving problems. First step to solving problems is to identify them.

> At some point, execution speed starts to matter more than the elegance of the code.

Also: at some point the elegance of the code starts to matter more than execution speed. :)

What surprises me most is some of the virulent reactions that code generation appears to elicit, sometimes citing reasons such as craft, artistry, and originality. As if the entire disciplines of computer science and systems engineering never depended on assemblers, code generation, compilers, JIT. Or really, just writing bytes that can represent machine code, P-code, or bytecode.

A reaction that doesn't appear to make the very direct connection with the systems of exploitation, but chooses to target the tools, or the users of tools is difficult to justify as extremely sophisticated.

I use AI tools daily and find them genuinely useful.

However I am increasingly annoyed at how everything has to be framed as a conversation about AI, how every tech-adjacent company has to brand itself as AI-first, and most of all, how overblown predictions are about an LLM being conscious, etc.

In short – it’s a useful technology reshaping tons of industries, but the hype is grating.

Because the HN crowd is composed largely of developers — the profession that is first to fall to the Axe of AI.

HN probably has as much as 5 million monthly users. This is not just a small group of insiders, but more of a broadly representative sample of startup and engineering people.

So there is a wide range of judgement, and more importantly, a diverse set of worldviews. These are beliefs that form the foundations of cognition and perception. In the general population there are a massive number of people who do not understand technology and/or do not really appreciate it at a deep level. This includes a significant percentage of startup and engineering people unfortunately.

I have loved using AI technology for 45 years (symbolic AI, old fashioned NNs, … to the present). I am also skeptical about the apparently desperation-driven ‘bet the farm’ approach we are taking here in the USA.

Slow is Fast.

Its complicated, this is how its going to be. People are going to have opinions and take sides or take no side at all.

I despise code written by VI! The only code anyone should ever run is code written with EMACS. With SPACES, not tabs. Because tabs take jobs from space bar pressers, and boil the oceans.

> Let’s face it: by the time I manually ship version 1.0 of a product, the AI-assisted version could have been deployed 10x faster.

I just don't understand what you mean by "let's face it". I repeatedly face it at my job, all our code has been AI assisted since March, and not once have I observed such a 10x speedup. The only 10x examples I've seen in the wild have been on tasks like cross-language code rewrites that completely fail your "code is just a means to an end" criterion.

Generally speaking the local crowd is anti-hype, and so it's easy to get the manifestation of that conflated with with what you're describing.

(I fit your literal description, but primarily from a nomenclature perspective - I'll call them generative models and LLMs - and I appreciate this puts me in the minority. BUT I do believe part of the hype feedback loop was the intentional mislabelling of these technologies from the outset. AND I understand why the marketers did that.)

I suspect the older crowd has lived through the hype playbook enough to recognise it early, and that the pattern this time around is becoming a bit a bit more obvious now, so I expect increasing levelling out of expectations & understanding.

Why do workers dislike being charged to rent scab labor from their oppressors?

1. because they know better. You don't have to understand it, for them to be right.

This comes from their years of experience. When you also get those years of experience, you may come to similar understanding.

Software Engineering was inherently romantic before AI.

People took time to understand the inner world of computers. Some people built brilliant solutions that represented the finest examples of human ingenuity. Knowledge was impressive. Side projects were impressive. The right engineer in the right place could make or break a business. Any industry that operates like this, where human skill and intelligence is valued and a key component of the process is beautiful.

With AI, all that has been snuffed out. No one gives a fuck. There is no skill required now. Talking about code with humans is pointless, talk to your AI. The meritocracy is over, this industry will soon be all about who you know, not what you know. Fuck your resume, your list of skills and experiences are quaint. You really think anyone gives a shit about languages you know or how many features and products you shipped? Anyone can do that shit with a few prompts of an LLM. So how else will you get a job? Know someone? Blow someone? Just hope you win the random selection?

A lot of people aren’t against the AI tech itself, they are against how it will change the tech culture. The old world is gone and the new one looks like it sucks, many people just don’t realize it yet, they are slow boiling frogs. They have not yet experienced being unemployed in the AI era.

Personally I'm not anti-AI, I'm anti-Stupid.

I also don't consider LLM's to be AI. I put it in the same category as PageRank.

And in my experience with AI doing personal projects, 10x as fast probably understates it by at least an order of magnitude.

Because unlike crypto and other tech fads, it’s hostile:

- job losses are immediately associated with AI in news

- privacy invasions, AI profiling, AI aggregators, etc.

- annoyance, AI chat bubble, AI useless tech support, AI interviews, etc.

- bandwagon “wrappers”, you know, wrap gpt api in saas and try to sell it in subscriptions, flooding show HN

- slops, slops everywhere. Codes, graphics, you name it.

And a lot more. AI to tech world is what smartphones did to internet, flooding non technical people into technical people’s space and basically ruining the fun part. Additionally, it didn’t bring any substantial breakthrough, in the past 3 years or so, did we have any breakthrough innovation in any sector as a result of AI? Barely, so you end up with a lot of noise flooding the internet, bots now are more than humans.

> Users don’t care whether the code was written by AI or by hand, or which framework you used. They care that the product works.

If I wanted to care about what users want, I would have been a founder or salesman, not an engineer.

Many reasons. Each person is an individual and you will learn most by seeking to engage with all the individual stories to empathise and understand. Ultimately it’s a very human thing to care, this is a big change and everyone deals with change differently. It’s a great start to see you caring and wanting to know more.

It seems weird to be "pro-AI" or "anti-AI" in general. It's a tool. It's like saying construction workers are pro-hammer

If I was an end user of a working product (AI or not), I wouldn't care.

At work generating and fixing loads of slop is less rewarding work than doing old coding, troubleshooting, article writing, whatever. The internet is full of fake blogs full of fake information. Youtube is full of fake videos and people reading LLM scripts. It feels impossible to share or appreciate small projects because it's so much harder to tell if any effort or thought went into something at all now. My parents can't tell what's real on social media. I'm less sure in my career path because I might spend my time learning skills that become useless in 5 years. I have conversations on the internet or Jira where people respond with LLM output (half the time saying "Claude says..." half the time not.) Kids are cheating their way through school. I'm probably getting dumber by using it.

There's plenty of reasons to be "anti-AI". It's not just a tool that's making programming more convenient.

It's because this place is full of people that are the developer equivalent of someone that constantly tells everyone they drive a manual transmission vehicle and it makes them superior.

lazy so copying from a different thread:

you might be drinking some of that AI koolaid, conflating our suddenly hypertrophied abilities to produce code regardless of our familiarity with the syntax or the APIs with ability to produce and deliver good quality products, but this delusion is getting reality check as we speak.

a realization is propagating through the industry that being able to produce more code than you're able to review, comprehend and internalize is actually not a great thing.

https://news.ycombinator.com/item?id=48381598

that said im not anti ai, i just think it is being applied in the most moronic ways during this hype cycle (gary tan anybody?)

Many of the people reading HN will be making a lot less money in two years. Some will be unemployed. Some will be homeless.

Human intelligence becomes less valuable in quantity as AI gets better. Being big and strong was once valuable. Not so much any more.

"When this machine learns your job, what are you going to do?"

There are coders and creators. The first identify as their tools, the second don't care about the tools (too much).

This explains to me 90% of the reactions I get when I talk to people.

> I’ll probably make a lot of enemies by saying this, but do people realize that code is just a means to an end

The means to an end I care about is that writing code was a means for me to make a living

People can pontificate all they want about how software engineering was never really about writing lines of code and at some abstract level they are correct

Your average software engineer still spends (spent?) a lot of their day writing code, it is the activity that delivered the actual value of our skills!

How do I deliver value to keep earning that paycheck now? It has been massively undercut away from me by AI systems. I do not see a good future for myself anymore

Am I not supposed to feel so negatively about it?

More like, HN crowd is anti-HYPE

The other day, I had a similar thought about the relative importance of code.

I'm working on a game, and I've been fussing over the code quality. And yeah, having code that isn't awful is important for various reasons. But with a game, it got me thinking, the code is literally the only part of the game the player doesn't experience.

The time I'm spending on the code, I could be spending on the art, the game design, the music, the story...

But my natural tendency is to hyper-focus on the only part of the game nobody will ever see. I thought that was interesting.

(That being said the codebase is ass and I do need to clean it up!)

Because a lot of us are engineers. It's our mindset and our job to question hype and broad strokes and easy solutions, to go a few levels deeper and ask "okay but does it really work?". I don't think most people are anti AI more than they are anti any tool.