undefined | Better HN

story

0 pointszhobbs1y ago0 comments

>For what? There is no ROI at that price point. There is no monetisation potential.

I think your whole argument is based on this being true, but you didn't give much argument about why there is no ROI. 400M USD isn't hard to generate...even a moderate ad engagement lift on X would generate ROI and that's just 1 customer.

Imagine going back in time and showing every VC how great the search business will be in 20-30 years. The only rational response would be to make giant bets on 20 different Googles...and I think that's what's happening. These all seem like rational investments to me.

0 comments

makestuff1y ago

Ken Griffin had an interview where he said something along the lines of the technologies dot com bubble pretty much turned out to be what everyone thought they would become at the time. The issue was valuations grew way too fast and it took much longer than expected for the companies to build out their products.

I think a similar thing is playing out with AI. In 5-10 more years these LLMs will replace a google search today (and maybe be even better).

loandbehold1y ago

Everyone I know has already switched from Google to ChatGPT for most of their search queries.

gmerc1y ago

That's a red herring because it ignores the part where they could have done the same things spending a tiny fraction of the money.

gordonhart1y ago

_Could_ they have done the same thing with a tiny fraction of the money? Grok 3 benchmarks are SOTA for both base model and reasoning. By definition, nobody has been able to do the same thing with any amount of money (discounting o3 which has been teased but is unreleased). That may change in the future! But as of now this is the case.

gmerc1y ago

So apart from the part where SOTA doesn't mean anything in the real world (there is no monetisation, there's no moat), please, it's benchmarks, we all know how you beat those since 2023.

Time to review https://arxiv.org/abs/2309.08632 AI-CEO.org's best friend

(and actually o3-mini-high beat them in a bunch of benchmarks so they removed it from those charts in the livestream)

YetAnotherNick1y ago

Why don't you do it then? If you are talking about Deepseek "$5M", then you would be interested to know that they pay 7 digit salaries and reportedly have H100s worth $2B[1].

[1]: https://sherwood.news/tech/the-trillion-dollar-mystery-surro...

zhobbsOP1y ago

Just wonder if it matters? If Google spent 10x as much in the first 5 years of its life would it be a worse company now? Giant TAM, winner takes all (or most?), all that matters is winning.

loandbehold1y ago

People like Demis Hasabis and Derio Amodei say that R1 efficiency gains are exaggerated. $5M training cost seems to be fake as sources suggest they own more GPUs.

j / k navigate · click thread line to collapse

0 comments

makestuff1y ago

I think a similar thing is playing out with AI. In 5-10 more years these LLMs will replace a google search today (and maybe be even better).

loandbehold1y ago

Everyone I know has already switched from Google to ChatGPT for most of their search queries.

gmerc1y ago

That's a red herring because it ignores the part where they could have done the same things spending a tiny fraction of the money.

gordonhart1y ago

gmerc1y ago

So apart from the part where SOTA doesn't mean anything in the real world (there is no monetisation, there's no moat), please, it's benchmarks, we all know how you beat those since 2023.

Time to review https://arxiv.org/abs/2309.08632 AI-CEO.org's best friend

(and actually o3-mini-high beat them in a bunch of benchmarks so they removed it from those charts in the livestream)

YetAnotherNick1y ago

Why don't you do it then? If you are talking about Deepseek "$5M", then you would be interested to know that they pay 7 digit salaries and reportedly have H100s worth $2B[1].

[1]: https://sherwood.news/tech/the-trillion-dollar-mystery-surro...

zhobbsOP1y ago

Just wonder if it matters? If Google spent 10x as much in the first 5 years of its life would it be a worse company now? Giant TAM, winner takes all (or most?), all that matters is winning.

loandbehold1y ago

People like Demis Hasabis and Derio Amodei say that R1 efficiency gains are exaggerated. $5M training cost seems to be fake as sources suggest they own more GPUs.

j / k navigate · click thread line to collapse