Hacker News Reader

Comment by hmokiguess | Hacker News Reader

While this article is just about optics, I would say the comments here about how the coding agents fare fail to realize we’re just a niche when compared to the actual consumer product that is the Chatbots for the average user.

My mom is not gonna use Claude Code, it doesn’t matter to her. We, on Hacker News, don’t represent the general population.

View on HN

Claude Code purportedly has over a billion dollars in revenue.

In terms of economic value, coding agents are definitely one of the top-line uses of LLMs.

Sure, I don’t disagree, but a fact remains that 1B is less than 10% of OpenAI’s revenue with ChatGPT and its 700M+ user base.

Coding agents are important, they matter, my comment is that this article isn’t about that, it’s about the other side of the market.

And OprnAI will never be worth its current valuation or be able to keep its spending commitments based on $20/month subscriptions

Claude has been measurably worse over other models, in my experience. This alone makes me doubt the number. That and Anthropic has not released official public financial statements, so I'll just assume it's the same kind of hand waving heavily leveraged companies tend to do.

I actually for for ChatGPT and my company pays for Copilot (which is meh).

Edit: Given other community opinions, I don't feel I'm saying anything controversial. I have noted HN readers tend to be overly bullish on it for some reason.

That doesn’t reflect my (I would say extensive) experience at this point, nor does it reflect the benchmarks. (I realize benchmarks have issues.)

Are you using Claude as an agent in VSCode or via Claude Code, or are you asking questions in the web interface? I find Claude is the best model when it’s working with a strongly typed language with a verbose linter and compiler. It excels with Go and TypeScript in Cursor.

I have used it for GDScript, C++, Java, and other more general questions. Specifically, comparing it to other LLMs responses ESPECIALLY after incremental narrowing by prompt. Claud seems to randomly change approaches and even ignore context to the point you get the same circular issues you see in Copilot (do A because B is bad, then do B because A is bad or worse ignore everything before and do C because it's nominal). It seems more primitive in my sessions from the last time I used it (for a couple days) ~45 days ago.

Reminder that the entire AI industry is loaning itself money to boost revenue.

I seriously question any revenue figures that tech companies are reporting right now. Nobody should be believing anything they say at this time. Fraud is rampant and regulation is non-existent.

On a purely theoretical-finance level, I don't think the circular funding is actually a problem in itself. It's analogous to fractional reserve banking.

Whether there's also fraud, misreporting of revenue, or other misbehaviour of weird and wonderful classifications that will keep economics history professors in papers for decades is a separate question. I just find that people get fixated on this one structural feature and I think it's a distraction. It might be smoke, but it's not the fire.

Doesn't fractional reserve banking depend upon independence of the various customers? The widely-reported circular financing between AI players does not enjoy that.

Anyone can sell dollar bills for 90 cents. When they can actually make a profit, then it will be impressive.

But that’s not what they’re doing, sir.

Are they profitable?

No, but that is not the same thing as selling dollar bills for 90 cents.

My mom uses the Google app instead of just going to Google.com on Safari. She’s probably going to use Gemini because she’s locked into that ecosystem. I suspect most people are going to stick with what they use because like you said, to consumers, they can’t really tell the difference between each model. They might get a 5% better answer, but is that worth the switching costs? Probably not.

That’s why you see people here mention Claude Code or other CLI where Gemini has always fallen short. Because to us, we see more than a 5% difference between these models and switching between these models is easy if you’re using a CLI.

It’s also why this article is generated hype. If Gemini was really giving the average consumer better answers, people would be switching from ChatGPT to Gemini but that’s not happening.

The other issue with this is that AI is still unprofitable and a money hole.

If consumers refuse to pay for it, let alone more than $20 for it, coding agent costs could explode. Agent revenue isn’t nearly enough to keep the system running while simultaneously being very demanding.

AI development is a money pit, AI use is profitable. Average ChatGPT subscribers are using way less than $20 of electricity and GPU time per month.

And it will get even more profitable for free users when ads roll in.

When you take depreciation into account, it's probably less profitable than a government bond.

Will your mom pay for chatgpt or just stop when they will try to start converting more users ?

Anecdotes, etc. but my 65 year old dad is pretty low tech and he was paying OpenAI $20/month before I was.

Little off topic but I just got done cleaning up my friend's dad's estate. He had dementia the last ~5 years of his life.

The amount of random fucking subscriptions this senile old dude was paying is mind boggling. We're talking nearly $10k/month in random shit. Monthly lingerie subscription? Yup, 62 year old dude. Dick pill subscription? Yup. Subscription to pay his subscriptions? Yup.

It makes me really wonder how much of the US economy is just old senile people paying for shit they don't realize.

We also found millions in random accounts all over the place. It's just mind boggling.

He had dementia that bad at 57?

Ouch, that's not nice. My grandmother has been in care since 2020 and no idea who anyone is (forgot kids, husbands, etc), but at least she was in her 90s when it started going bad.

Also anecdote but most low-tech people I know are using chatGPT like google search and will never pay for it. Maybe that’s why chatGPT Ads will work beautifully with them

Everyone says “advertise!” Like it’s a magic bullet. The tech industry is littered with companies that have high traffic and couldn’t figure out how to monetize via advertising. Yahoo being the canonical example.

Besides the cost of serving an LLM request - using someone else’s infrastructure and someone else’s search engine is magnitudes higher than a Google search.

Besides defaults matter. Google is the default search engine for every mobile phone outside of China and the three browsers with the most market share

Ads per se aren’t necessarily a problem. It’s good to be able to discover companies et products. The problem is bad ads, ads that lie, ads for fake products or unfair ads. Also, too much ads is also a problem

Ads make the user experience worse. I don’t use any ad supported products - unfortunately outside of LinkedIn.

I 1000% agree with you. Companies are unreasonable and it kills the UX in a majority of cases. That makes us hate ads

> Everyone says “advertise!” Like it’s a magic bullet. The tech industry is littered with companies that have high traffic and couldn’t figure out how to monetize via advertising. Yahoo being the canonical example.

That is obviously because they can't figure out any other value of the product that's sellable.

Oooh, sick burn

Coding agents seem the most likely to become general purpose agents that everyone uses eventually for daily work. They have the most mature and comprehensive capability around tool use, especially on the filesystem, but also in opening browsers, searching the web, running programs (via command line), etc. Their current limitation for widespread usage is UX and security, but at least for the latter, that's being worked on

I just helped a non-technical friend install one of these coding agents, because its the best way to use an AI model today that can do more than give him answers to questions

AI coding has massive factors that should make it the easiest to drive adoption and monetize.

The biggest is FOMO. So many orgs have a principle-agent problem where execs are buying AI for their whole org, regardless of value. This is easier revenue than nickle-and-diming individuals.

The second factor is the circular tech economy. Everyone knows everyone, everyone is buying from everyone, it's the same dollar changing hands back and forth.

Finally, AI coding should be able to produce concrete value. If an AI makes code that compiles and solves a problem it should have some value. By comparison, if your product is _writing_, AI writing is kind of bullshit.

> If an AI makes code that compiles and solves a problem it should have some value

Depends if the cost to weed out the new problems it introduces outweighs the value of the problems solved.

To be clear this is me making the most generous case for LLMs, which is that some people really do just want a shitty app to check a box. In my experience fixing LLM-produced software is worse than just writing it from scratch.

I think LLM writing replacing actual authors or AI "art" is fundamentally worthless though, so at least coding is worth more than "worthless"

I've got to wonder what the potential market size is for AI driven software development.

I'd have to guess that competition and efficiency gains will reduce the cost of AI coding tools, but for now we've got $100 or $200/mo premium plans for things like Claude Code (although some users may exceed this and pay more), call it $1-2K/yr per developer, and in the US there are apparently about 1M developers, so even with a 100% adoption rate that's only $1-2B revenue spread across all providers for the US market.... a drop in the bucket for a company like Google, and hardly enough to create a sane Price-to-Sales ratio for companies like OpenAI or Anthropic given their sky-high valuations.

Corporate API usage seems to have potential to be higher (not capped by a fixed size user base), but hard to estimate what that might be.

ChatBots don't seem to be viable for long-term revenue, at least not from consumers, since it seems we'll always have things like Google "AI Mode" available for free.

Reply on news.ycombinator.com