Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

OpenAI raises $110B on $730B pre-money valuation

techcrunch.com

Post reply on HN
Print view View original

401–410 of 620 posts

Re: OpenAI raises $110B on $730B pre-money valuation

#401

Post by richardw »

beernet wrote: post #327
bentt wrote: post #318

Someone please explain how OpenAI is not Netscape 2026. They had first mover advantage but no network effect, no moat, and are racing to stay ahead of infinitely resourced incumbents.

How are ~1B active users not "moat"? Might have to pull out the "Haters gonna hate" like it's 2007

“In December, Gemini traffic increased by 28.4% month-over-month, while ChatGPT traffic decreased by 5.6%”

https://www.businessinsider.com/openai-chatgpt-vs-gemini-web...

"What's you number one piece of hiring advice?"

"Hire for slope, not Y-intercept. This is actually my number one piece of life advice."

-@sama, who I’m generally a big fan of. But the job is now harder

Re: OpenAI raises $110B on $730B pre-money valuation

#402

Post by XenophileJKO »

christoff12 wrote: post #392
onlyrealcuzzo wrote: post #325

Earlier quoted context omitted.

> each model is roughly 2x profitable on its own, but each next model costs 10x the last. The whole thing only works if scaling keeps delivering. This is a decent argument, but it's not the death knell you think. Models are getting 99% more efficient every 3 years - to get the same amount of output, combined with hardware and (mostly) software upgrades - you can use 99% less power. The number of applications where AI…

> Models are getting 99% more efficient every 3 years - to get the same amount of output, combined with hardware and (mostly) software upgrades - you can use 99% less power. Even if true, this still doesn't bend the curve when paying for the next model. > If AI stopped progressing today, it would take probably a decade or longer for us to take full advantage of it. So there is tons of forward looking revenue that isn…

I would actually like to see the real math currently.

The market adoption has increased a lot. The cost to serve has come down a lot per token.

Model sizes have not increased exponentially recently (The high point being the aborted GPT-4.5), most refinement recently seems to be extending training on relatively smaller models.

When you take this into account together, the relative training to inference income/cost ratio likely has actually changed dramatically.

Re: OpenAI raises $110B on $730B pre-money valuation

#403

Post by robotpepi »

onlyrealcuzzo wrote: post #325
7777777phil wrote: post #287

$730B pre-money for a company where each model is roughly 2x profitable on its own, but each next model costs 10x the last. The whole thing only works if scaling keeps delivering. Research (Sara Hooker et al.) is not encouraging on that front, compact models already outperform massive predecessors on downstream tasks while scaling laws only predict pre-training loss reliably. Wrote about both the per-model math and t…

> each model is roughly 2x profitable on its own, but each next model costs 10x the last. The whole thing only works if scaling keeps delivering. This is a decent argument, but it's not the death knell you think. Models are getting 99% more efficient every 3 years - to get the same amount of output, combined with hardware and (mostly) software upgrades - you can use 99% less power. The number of applications where AI…

ok, but everything depends on your numbers being correct. 99% improved efficiency seems kind of a way too optimistic prediction.

Re: OpenAI raises $110B on $730B pre-money valuation

#404

Post by davemp »

Amazon hedging $35 billion on an IPO (we know the odds of AGI with LLMs are vanishing small) is concerning. The hedge signals that they think it's a good "greater fool" investment, otherwise they should be happy to take equity in a company they think will be stable/profitable.

Re: OpenAI raises $110B on $730B pre-money valuation

#405

Post by swingboy »

sigmoid10 wrote: post #400
blmarket wrote: post #343

Earlier quoted context omitted.

> 99% more efficient every 3 years It's 2x efficiency. Then I'd take 50% less power instead of ridiculous 99% less power.

GPT-4 came out 3 years ago and you can run comparable models for 1% of the cost nowadays. That is not 2x efficiency. That's two orders of magnitude in end-to-end compute efficiency.

How do we know how much it costs? Or is this just based off the token pricing?

Re: OpenAI raises $110B on $730B pre-money valuation

#406

Post by recursive »

dminik wrote: post #382
max51 wrote: post #250

Earlier quoted context omitted.

"I give you 30B$ worth of hardware that costs me <10B$ to make in exchange for 30B$ worth of shares in your company" would be a more accurate description.

How does this work with private companies? It feels like Nvidia could find that the market does not value OpenAI stock the same way they did.

Public or private, there's never a guarantee of being able to sell back for some nominal price.

Re: OpenAI raises $110B on $730B pre-money valuation

#407

Post by danielparsons »

sigmoid10 wrote: post #400
blmarket wrote: post #343

Earlier quoted context omitted.

> 99% more efficient every 3 years It's 2x efficiency. Then I'd take 50% less power instead of ridiculous 99% less power.

GPT-4 came out 3 years ago and you can run comparable models for 1% of the cost nowadays. That is not 2x efficiency. That's two orders of magnitude in end-to-end compute efficiency.

you're looking at nearly the entire curve of the tech's development. that's like saying lightbulbs became 99% more energy efficient and therefore will become another 99% more energy efficient. but most techs follow an S curve.

Re: OpenAI raises $110B on $730B pre-money valuation

#408

Post by mcbits »

These are geopolitical activities akin to the arms buildup of the Cold War, but happening somewhat more openly through the guise of private sector investments. The names of the companies involved are ephemeral. The numbers "invested" are largely imaginary, just play money pumped into the system over the last few years, flowing around, searching for a place to park. The result will be a certain amount of tangible infrastructure in the form of datacenters, power plants, semiconductor fabs. It's a Hail Mary move to keep pace with a certain geopolitical competitor. This process may propel society to the next level or may collapse it, depending on how society chooses to use these resources and how successful the competitor is at its own similar endeavor.

Re: OpenAI raises $110B on $730B pre-money valuation

#409

Post by lysace »

max51 wrote: post #242
_fat_santa wrote: post #95

IMO this looks largely like another circular investment. Amazon's investment is tied to OpenAI using AWS for their Frontier product and I assume Nvidia's conditions are that OpenAI continue buying hardware from them. Then there's SoftBank though given that those are the same guys that invested heavily in WeWork, I assume this is just very brash bullishness on their part. From my perspective, I hope that OpenAI surviv…

The "circular investment" is mostly start up companies using their stocks instead of cash to pay for server hardware and cloud computing. There is a few extra steps in between that make things look weird and convoluted, but the end results is really just big companies giving hardware and getting shares of ai companies in exchange for it.

Cisco did this in 1999. That's how my smallish apartment building in Sweden ended up with a kick-ass Cisco 10 Gbps switch in its basement a year later - when these cost real money.

I think the HOA still only pays like $10/month/apartment for an entry level that's now defined as 250/250 Mbit/s. Someone must have been unusually savvy with the contracts.

https://newsroom.cisco.com/c/r/newsroom/en/us/a/y1999/m11/ci...

Cisco survived but it took them until late last year to recover their 1999 stock value (that's 26 years).

Re: OpenAI raises $110B on $730B pre-money valuation

#410

Post by danielparsons »

mlinsey wrote: post #348
monooso wrote: post #333

Earlier quoted context omitted.

Not GP, and not saying I agree with them, but it may be worth remembering that Netscape had 90% market share at one point. Active user count may not be the moat you imagine.

Adoption of web browsers was also much lower when Netscape was dominant. 90% marketshare is less meaningful if you're only 1% of the way to the potential market size. Peeling away users who talk to ChatGPT every day is very possible, but harder than getting someone whose never used an LLM before (but does use your OS, browser, phone...) to try yours first. I think the even better analogy than browsers is search engin…

google search definitely has a moat. people build their websites to optimize for google's algorithm, therefore google users see better results -> google gets more users -> websites optimize for google -> repeat. Personally I never bother with 'bing SEO' or 'bing ppc ads'.
Post reply on HN