Hacker News Reader

Comment by smashed | Hacker News Reader

If these numbers are right, it's actually not that bad. Cut r&d costs and they are mostly profitable.

Yes if you ignore all the reasons why they’re horribly unprofitable, they’re profitable.

R&D costs are hurting profit side and while you can cut that one just becomes irrelevant overnight in this space if you do, hence the problem.

> R&D costs are hurting profit

That’s quite the hot take, considering it’s literally an R&D company that got to where it is by doing R&D.

Isn’t the post above saying the same thing after the part where you cut it off…?

> R&D costs are hurting profit side and while you can cut that one just becomes irrelevant overnight in this space if you do, hence the problem.

I see where you're coming from but I still make a distinction between "R&D costs hurt profits but are necessary to stay relevant" and "R&D costs money but is the source of future profits". It's the difference between a necessary expense and an investment.

Not to mention they will need to research how to make their models faster and cheaper to run in order to fit some margin within what people are actually willing to pay.

So you’re saying if you cut all the cost centers a company would only have profit centers? If you ignore all the losses you’ll only have profits?

It's more like once you figure out how to make a really good lamp then producing lots of lamps will be profitable. But the lamps are currently suboptimal so we'll be in the red until that time.

And then someone will come up with lamp pro max and you’ll be out of business. You realize why R&D exists in tech companies even though it’s a cost center right?

OpenAI won't be able to cut R&D spend and collect rent on their existing models as long as the Chinese models keep up the pace of being ~6 months behind them for a fraction of the price.

And if you wait 12 months, someone will be giving away lamps for free that work just as well.

It's more like you have a business making engines, each generation of engine has eventually turned out to be profitable over its lifespan, but each generation has an exponentially increasing R&D cost and your customers will switch from the old engines to a competitor if they don't like the newest generation.

You're stuck racing against your competitors with the distinct possibility that your R&D costs will outgrow the market demand, and you can't stop because otherwise your customers will stop investing in your dead end tech and switch.

Except this is the first generation of engine manufacturers and nobody knows if it will actually be profitable yet.

We do know that Anthropic claims earlier models eventually turned a profit, and OpenAI is presumably the same.

What is in doubt is whether past performance is an indicator of future results. How long will the ever-increasing R&D expenditure keep paying off?

And there are just tons of free engines sitting around that are basically almost as good as the newest ones...

Except that your ability to sell that lamp is conditioned on other people not giving away the same lamp for free. GLM 5.2 is free.. which leaves OpenAI with what? Harness layers that competitors do better? Seems like OpenAI needs to keep training, but distillers will always be able to distill cheaper than you can train.

High risk high reward I guess

This is private equity 101 no?

OpenAI can easily cut R&D costs by replacing engineers with Claude Code

I am having difficulty parsing this sentence ... :-)

Cut down on the one thing they need to keep themselves relevant in this space?

Watch them flare out like a star… but there is lots of questions re the the return on RnD. Is it worth spending another order of magnitude for only marginal frontier gains?

Actually reduce R&D to ZERO and they are still losing money.

While you cant discount 100% R&D they are close, agreed

I bet any FAANG spend is mostly R&D.

If it's not materials, not energy or taxes, not manufacturing, not licensing or rental fees, then I can only think of R&D.

People keep overlooking the fact that costs for these providers scale along with customer acquisition. Most startups don't have that linear expense. Also, training costs are accelerating to get new models out faster. One doesn't simply "get rid of R&D" costs as a comment upstream mentioned. I can't actually imagine R&D goes down anytime soon unless you're willing to play third fiddle.

Unless these frontier providers feel some type of squeeze or constraint the Chinese are well positioned to leave the US bag holders of an NVidia bound system. And if anyone has to wonder how one provider for a critical piece of infrastructure will go, well...

If they cut down on R&D they will be no better than the open source models you can run at cost yourself.

Even if they keep the R&D costs, more efficient inference and 0 Marketing spend also gets you there. Inference is honestly super inefficient at this point, we can do far better than GPUs, push utilisation up, build more efficient datacentres.

Numbers are probably not right as classifying everything aa r&d is going to the temptation

Reply on news.ycombinator.com