AI will end up being a commodity. That's also why you already see the first rounds of protectionism under the guise of national security. Obviously they do have a point but that point works against them just as easily. Anybody that integrates unreviewed model output into their workflow is asking for trouble.

> There has always been a lot of evidence that prices are only going down.

And with it, the cost of intelligence / digital services / much of white collar work is going to zero.

It’s helpful for a model to be somewhat static when it’s involved in business processes.

They need to be retrained often is my experience for general usage.

I have SO many cases, for example when shopping, when they make assumptions based on their cutoff date, and they need to redo everything based on lots of searches.

Sure, for many use cases old models will be useful but for general usage a model that's 5 years old will probably be garbage for most people, unless they prompt it extensively, which the general public won't do.

> But I'm also interested is retraining needed? how relevant will a model be in N years if it's just "static", and if you do need to retrain and fine-tune over time, is it still profitable? how often is that needed?

Just wait until what Ilya Sutskever has for Safe Super Intelligence (SSI) to self update the weights without retraining them to bring training costs down close to $0.

Nobody meets expectations of a 2 trillion dollar IPO selling a fungible commodity like tokens for old models.

And that's the rub - it doesn't matter if inference is profitable, the US labs have promised the whole kitchen sink, and taken serious investment on that pitch. They won't make it back in a cut-throat market for inference. They have to continue pitching the idea that only their models are worth using, for quality/safety/performance/etc reasons.

Which means continuing to develop and release new models. At huge expense.

Side note - while deepseek seems to be doing fine, I'll note that the emphasis on cost effectiveness is very much not a thing for the US labs right now... and it was for non-us labs. I'd be quite interested in knowing if anthropic is pulling in the same returns, or what the margin on their subscription plans actually is.

---

I'm already seeing this directly at work - lots of devs still happily use claude for development, but the products we're making aren't. They'll use kimi, glm, gpt-oss, etc... basically - if cloudflare is hosting it, we're exploring it (mainly for contractual/security reasons): https://developers.cloudflare.com/workers-ai/models/.

Because we are price sensitive, and we already have to do all the work of creating very robust evaluation frameworks for our products, so we might as well choose the cheapest model that meets our standards.

whiiiiich is why we're seeing things like Haiku and gpt-luna priced down near the open models at ~.10/mm in, ~.50/mm out. They've been forced to cut prices to keep commercial customers on their cheaper models instead of moving away.

This will continue. If inference is profitable and easy, prices will continue to erode. Go browse openrouter, plenty of providers happy to undercut the competition. Bigger shops with auditing requirements will avoid them, but smaller ones won't.

I assume they need these models to stop being replaced in 1-2 months though.

Would that be due to the size of the models from DeepSeek or perhaps only for their distilled models? If it's for general purpose models that would crash the valuation of Anthropic and OpenAI, both who are struggling with profitabilitet to say the least.

It's really hard to unpick, for example, energy prices might be heavily subsidized for the Chinese models.

But it's clear that it's not the money burning pit I think it was just a year ago.

> Whereas if you're handing off all dev and no longer reading code, just making CTO-level architecture decisions, then there are only 2~3 models right now worth using, none cheap.

I use frontier models every day and night, but not sure if the described approach is possible with any of them right now. Models still make stupid mistakes and people who do not read the code produced are bound for surprising discoveries in their codebase. Not trying to downplay the effectiveness of LLM-assisted coding and all that, but it's hard to imagine how much stupid shit i would've committed if I was not reading their output.

Only if CPM can be high enough to counter balance the inference costs.

Current CPM is between $30 and $60 which is x20 to x30 more expensive than alternatives.

Now we have to wait and see if performance is also at least 30x of previous ad targeting.

The thing that stops it is, ironically, personal agents. Get one to interact with the ad-driven one if you need to and filter ads out.

Artificial intelligence will 'infect' every surface of life and business. The Internet is merely a distribution platform for intelligence (and yes that includes TikTok videos). Its first distribution chore was slow and quaint, it was human. The intelligence was always the higher value, not the distribution network.

Distribution of intelligence: data, photos, video, communications, transactions, entertainment, et al.

And to be clear, it doesn't matter if some random commentors on HN don't think very highly of some or all of the intelligence being distributed. Their elitist opinion of the quality of intelligence of a given thing doesn't matter as to whether it's happening or not.

When a 16 year old records a cloned TikTok dance video, edits it, and then publishes it, that's intelligence distribution. That's a person thinking through a series of actions to try to communicate to an end audience, via an entertaining medium. That's intelligence being distributed over the Internet. That's the quaint, slow, human form of it.

Up to this point that distribution was basically entirely human, capped (post saturation) at slow growth N intelligence output (derived from the total number of humans and their time / productivity capabilities). AI uncaps that and radically expands intelligence output & distribution at a growth rate that is so fast you might as well start picturing the planet having tens or hundreds of billions of people on it. The human population will shrink, the artificial population will increase dramatically. People will have more artificial friends than human friends within ~20-30 years.

A few decades from now the notion of thinking of computing separate from AI will be so foreign most people won't understand what you're talking about. It will just be thought of as computing. Humans will forget a time when we weren't augmented by the constant presence of artificial intelligence. Try taking 'smart' phones away from the masses. In about a decade try taking away their helper agents. A decade after that try taking away their home robots.

The steroid jump is AI + robotics. The Internet is trite compared to what comes out of that over the rest of this century. Robotics will physically remake our world, in a way the Internet can't. We'll deposit tens of billions of robots onto the surface of the planet and they will almost all be the result of having figured out how to make intelligence (relatively) easy & (relatively) cheap. I'm not talking about Roombas.

The Information Age is over. We're into the Age of Intelligence (insert the unavoidable quips about humanity & intelligence).

The Internet is just one way to move intelligence around (a robot body is another, a smartphone is another, a human body is another, and so on). It's a sub value to intelligence broadly. Intelligence is the ultimate value. The valuations (and sales growth) for OpenAI and Anthropic is because they're vast intelligence output machines in the Age of Intelligence, ditto Nvidia's business boom. We can now mass manufacture intelligence at an accelerating clip, and it has gotten cheaper at a comical rate over the past few years. While HN comments snort about pelican improvements day to day, the leap over a few years is astronomical in regards to what significant AI systems can now do (with LLMs being just one type of that).

> actively trying to prevent buildout of solar and wind power.

And actually paying people not to ! (Paying RWE $1.2bn to abandon wind contracts[1])

Its absolutely nuts. Pure generational can-kicking.

[1] https://www.bbc.co.uk/news/articles/c1e1vg0gjl5o

Why? AI DCs can't rely on solar or wind so there must always be a thermal backstop. China is building lots of coal and nuclear.

> Huge, generational mistake for the USA to be leaning so hard into fossil fuels right now and actively trying to prevent buildout of solar and wind power.

If the US is trying that, they are failing pretty hard - 88% of new power in the US in 2025 was renewables, and 2026 is estimated to be above 90%

One of countless mistakes made by the current administration. Decimating research funding and firing many scientists is another.

It's hilarious and depressing that one of Trump's supposed goals was making the US into an industrial powerhouse again, and yet he's acting to protect the profits of the fossil fuel industry as opposed to encouraging the kind of resource efficiency that would make the US competitive long term.

That would depend on whether you are a user/customer or an investor/owner in AI. At the moment, nearly everyone is on the investor/owner side. The entire western world has doubled down on AI becoming the most popular product since the invention of fire. If it turns out that fire 2.0 cannot be patented/copyrighted/licensed for insane profits, those dollars disappear.

Depends if you’re a shareholder or not

The tech is at the point where synthetic data is almost high enough quality to train further models on, AND there are so many human users of cloud AI, that their usage can be gathered and trained upon (it's human generated and used in real world situations, so, it is of a good enough quality). The cat is fully out of the bag.

The upstream models weren't built by god, they were built by distilling the world's knowledge and cultural products.

If OpenAI/Anthropic stop doing it there's no reason to believe someone else won't. Or that we won't come up with a different approach.

Anthropic is zooming to $100 billion in annualized revenue.

The tremendous cost is matched by the tremendous business output.

In the span of less than five years, starting from their first $1 of revenue, Anthropic + OpenAI will have produced businesses doing at least $150 billion in combined sales. The demand isn't reducing at present, it's climbing, they can't come close to servicing the total demand for their frontier models.

SAP is under $40 billion in annualized revenue. They're 54 years old.

All LLM are subsidized and blatant bubble.

If the only profit is from running commodity free software (as OP claims), there's no moat and OpenAI/Anthropic are worthless.