Live data from Hacker News

I think Anthropic and OpenAI have found product-market fit

simonwillison.net

661–670 of 1001 posts

Re: I think Anthropic and OpenAI have found product-market fit

#661

They've got, ballpark, $5t to $10t to make back in the next 5 years, or the hardware buildouts will start getting written down. This means we're going to need $1t+ per year in spending, per year, on tokens. 200m knowledge workers in the world, 30m developers. We're talking about a world where you need 5% of every knowledge workers salary to go into tokens. 20% if you're a developer. That's a _huge_ shift. Most people…

I work for a tiny little company ($150MM annual rev with 9% net) and we are already looking at dropping $100k on hardware to run local models because, for us, they're "good enough." Our estimated spend for AIaaS would exceed that cost in less than a year. In a few years, there will be hardware capable of running frontier models good enough for most things at accessible prices for even tiny companies.

You people are delusional. How many times a day am I going to read this fiction of "good enough in a few years for most things".

There are physical limits to how much you can compress data and how much is needed for a capable model. If by hardware capable for running SOTA you mean a 7 figure investment for a company, than sure. But how come these companies didnt do the same thing for cloud? There's been this option for self hosting infrastructure for a decade but companies don't use it, they pay AWS.

Re: I think Anthropic and OpenAI have found product-market fit

#662

Earlier quoted context omitted.

I'm about to leave a shallow comment, but I am a bit skeptical of the supposed drop in inference costs. If AI labs saw a lot of potential there, they'd surely be bragging about it non-stop? So the fact that publicly available information is conflicted is probably a sign that at the very least, the numbers aren't amazing. Yes I know there's no evidence and this is lazy reasoning. But there's probably a bit of truth to…

Why on earth would AI labs be bragging about how little the product they sell actually costs them to make? You don't want to do anything that reduces it's perceived value to the user, that might make them less willing to pay for it. Also, inference costs are bound to go way down with more optimized architectures. GPUs are fundamentally not great at inference. No platform where the weights are streamed from a large po…

Because the most important thing for any pure play AI company right now is to prove they are a viable company. And sure they have proved they can make billions, but also that they can lose billions more. They are going to need even more money and to prove to the next round of investors at an even higher valuation that they are a viable business they need to show not that they can generate revenue, but that they can one day turn a healthy profit. And that is the trillion dollar question.

Re: I think Anthropic and OpenAI have found product-market fit

#663
post #631

I have to give them kudos. This whole thing is the greatest swindle of all time. AI has some use cases, but not at the price it’s currently priced at. I’ve been on AI since GPT-2 with a lot of heavy users. Every user has the same story, curiosity, surprise, hype, hate, realization. Enterprise is usually a bit behind and are right now at hype cycle, that’s where they sold all the deals and do the IPO. It’s really a VC…

Trying to parse the raw claims here maybe you can help me out - it’s a swindle because ROI of tokens for coding models is not positive? As in it doesn’t bring enough value to charge like the $100/mo? - enterprise customers are too dumb to see this - IPO to max out the CEO profits for what is ultimately blockchain vaporware Am I getting that right? Or am I putting words in your mouth? > it gives tons of people in the…

> Dario and Sam are saying “if you buy our coding agent subscription you can build a game with zero skill and one shot and then be rich”?

They are saying profitable companies should replace the engineers that built their systems with a subscription (while they are hiring).

Re: I think Anthropic and OpenAI have found product-market fit

#665
post #575

Earlier quoted context omitted.

We don't have to take Ed's word for it. Anybody who's capable of doing grade school math can see that the numbers simply don't work. These companies are literally spending orders of magnitude more money than they're actually bringing in. Cursor, who've been renting Claude, estimated just recently that a $200-per-month Claude Code subscription could use up to $2,000 in compute. https://www.forbes.com/sites/annatong/20…

Interesting story. Here's what it says: > According to a person familiar with the company’s internal analysis, Cursor estimated last year that a $200-per-month Claude Code subscription could use up to $2,000 in compute, suggesting significant subsidization by Anthropic. Today, that subsidization appears to be even more aggressive, with that $200 plan able to consume about $5,000 in compute, according to a different p…

That's the big question, and nobody really knows what the operating costs actually are right now.

Re: I think Anthropic and OpenAI have found product-market fit

#666

They've got, ballpark, $5t to $10t to make back in the next 5 years, or the hardware buildouts will start getting written down. This means we're going to need $1t+ per year in spending, per year, on tokens. 200m knowledge workers in the world, 30m developers. We're talking about a world where you need 5% of every knowledge workers salary to go into tokens. 20% if you're a developer. That's a _huge_ shift. Most people…

> Most people I know cite +20%-40% velocity

Seems roughly right, that does seem to be about the boost in the most well-suited cases where you essentially know exactly how to solve the problem, the problem won't change much, and it's truly a matter of just churning out the implementation.

In that case precisely prompting, doing the review & nudge loop, can be a pretty nice (nice, still not game changing) speed boost over literally typing out the code to match the design in your head.

The less optimistic view though is that most things you build aren't like that. Even if they seem like it first. These things get booked as a nice speed boost, but you'll only find out much later they weren't.

A confounding factor is that it seems like many people not in the detail of building software do seem to think of most to all things are like that, even before AI assisted coding. Not much need to say more - see the entire history of the 'agile' movement for evidence of this.

And because most things aren't like that, I actually struggle to see fundamentally how more than 20-40% will ever be achieved (short of the ever-present deus ex machina of AGI argument), simply because the generation is already really good for these types of things. So since things like this aren't going to increase in overall proportion of things to be done, I don't see where the overall extra gains come from by models improving at this point.

Re: I think Anthropic and OpenAI have found product-market fit

#667

They've got, ballpark, $5t to $10t to make back in the next 5 years, or the hardware buildouts will start getting written down. This means we're going to need $1t+ per year in spending, per year, on tokens. 200m knowledge workers in the world, 30m developers. We're talking about a world where you need 5% of every knowledge workers salary to go into tokens. 20% if you're a developer. That's a _huge_ shift. Most people…

Also hardware will be obsolete or dead in 5 years, and warrantys are 3 years from Nvidia. Ask crypto miners how these kind of hardware economics work. Numbers have to keep going up all around. Its a fundamentally broken business model unless prices increase 10x

Re: I think Anthropic and OpenAI have found product-market fit

#668

They've got, ballpark, $5t to $10t to make back in the next 5 years, or the hardware buildouts will start getting written down. This means we're going to need $1t+ per year in spending, per year, on tokens. 200m knowledge workers in the world, 30m developers. We're talking about a world where you need 5% of every knowledge workers salary to go into tokens. 20% if you're a developer. That's a _huge_ shift. Most people…

I work for a tiny little company ($150MM annual rev with 9% net) and we are already looking at dropping $100k on hardware to run local models because, for us, they're "good enough." Our estimated spend for AIaaS would exceed that cost in less than a year. In a few years, there will be hardware capable of running frontier models good enough for most things at accessible prices for even tiny companies.

> there will be hardware capable of running frontier models

The current frontier? Sure. The frontier then? No - obviously that frontier is going to keep consuming available datacenter compute capacity, which will be better

Re: I think Anthropic and OpenAI have found product-market fit

#669
post #571

Earlier quoted context omitted.

I just don’t understand how people are getting negative value out of AI or even only 20% productivity boost. I can only conclude that people don’t know how to use agents.

Are you mostly creating new things or integrating with complex, undocumented, untestable systems?

Mostly brownfield systems in Java, Elixir and TS. I use OpenSpec in explore mode and point the agent to all the different repositories (when not working in a monorepo) to identify changes. Once done, i switch to propose mode and spend at least 15 minutes there iterating over the plan until I'm satisfied with the TDD approach (agents need tests to verify their work). Then apply and review. This also auto generates docs etc.

Re: I think Anthropic and OpenAI have found product-market fit

#670
post #378

Earlier quoted context omitted.

And what was your return on investment?

As I commented elsewhere, I'm still bad at making money from my open source work: https://news.ycombinator.com/item?id=48296794#48298909 (I have a feeling if I could say "and I closed $2m in sales with the software I wrote!" people would find a way to say that didn't mean anything anyway, because how can I prove I wouldn't have made those sales writing it by hand?)

> "and I closed $2m in sales with the software I wrote!"

Given the audience you are reaching, that is actually the expectation. Github stars is not a great metric.

Post reply on HN