The guys comments on AGI do not follow scientific method - my hypothesis is Meta wants to have enough AGI influencers so that they can dictate definition as goalposts change, not quite based on skill set obv.
Meta Spends $14B to Hire a Single Guy
51–60 of 77 posts
Re: Meta Spends $14B to Hire a Single Guy
#52I'd push back on a couple things here. The notion that Scale AI's data is of secondary value to Wang seems wrong: data-labeling in the era of agentic RL is more sophisticated than the pejorative view of outsourcing mechanical turk work at slave wages to third world workers, it's about expert demonstrations and work flows, the shape of which are highly useful for deducing the sorts of RL environments frontier labs are…
- Could be wrong about Scale. I'm going off folks I know at client companies and at scale itself. - idk I've trained a lot of models in my time. It's true that there's an arcane art to training LLMs, but it's wrong that this is somehow unlearnable. If I can do it out of undergrad with no prior training and 3 months of slamming my head into a wall, so can others. (Large LLMs are imo not that much different from small…
Dude you went to Columbia, you probably dont think people that went to state schools are even human.
Rest of the article was good
Re: Meta Spends $14B to Hire a Single Guy
#53This post doesn't quite comprehend why Meta made a 49% investment instead of an acquisition. The path Meta chose avoided global regulatory review. FTC, DOJ, etc and their international counterparts could have chosen to review and block an outright acquisition. They have no authority to review a minority investment. Scale shareholders received a comparable financial outcome to an acquisition, and also avoided the regu…
I agree that this is a possible reason. Meta wants to move fast and m&a is too slow for their tastes. I make the case in the article though that the actual acquisition doesn't really make sense for metas core business, but I agree it's possible. I disagree that this is a win/win. Scale stock is still illiquid, and people who remain at scale or have scale stock are now stuck with shares that are actually less valuable…
Shareholders retain their stock, but also received a dividend equivalent to selling all their shares at a premium to the previous valuation. It is actually an incredible deal for them. (Source: I'm one of them.)
Re: Meta Spends $14B to Hire a Single Guy
#54This post doesn't quite comprehend why Meta made a 49% investment instead of an acquisition. The path Meta chose avoided global regulatory review. FTC, DOJ, etc and their international counterparts could have chosen to review and block an outright acquisition. They have no authority to review a minority investment. Scale shareholders received a comparable financial outcome to an acquisition, and also avoided the regu…
Right, Microsoft did this, Google did this for character AI.
I don’t really understand the purpose of all this, Scale is not an ai research lab, it’s basically Fiverr.
Why would bringing people over from there make Meta more compelling to AI researchers?
Re: Meta Spends $14B to Hire a Single Guy
#55Earlier quoted context omitted.
Same way companies fell behind during the original "dot com" era: By not burning money quickly enough. Disclosure: I work for one of the companies that fell behind.
Was that the real reason companues failed? By not burning money, instead of burning through it too fast?
Sorry for the late reply. I was being somewhat sarcastic, or maybe not. During the "dot com" era, investors were throwing money at anybody willing to take it, and a start-up was questioned about whether it was maintaining a sufficient "burn rate." In hindsight, it was a speculative bubble.
The company that I work for survived that era by not doing those things.
Re: Meta Spends $14B to Hire a Single Guy
#56Earlier quoted context omitted.
I agree that this is a possible reason. Meta wants to move fast and m&a is too slow for their tastes. I make the case in the article though that the actual acquisition doesn't really make sense for metas core business, but I agree it's possible. I disagree that this is a win/win. Scale stock is still illiquid, and people who remain at scale or have scale stock are now stuck with shares that are actually less valuable…
Shareholders retain their stock, but also received a dividend equivalent to selling all their shares at a premium to the previous valuation. It is actually an incredible deal for them. (Source: I'm one of them.)
O shit really? That's a massive update and possibly invalidates a lot of this article. It's also the first I've heard of this -- can you tell me more? Are you saying everyone essentially got bought out? (if you want to dm me in case its still pseudo-private, feel free to at theahura at gmail)
ETA: there's now an update header on the article based on this information
Re: Meta Spends $14B to Hire a Single Guy
#57Earlier quoted context omitted.
- Could be wrong about Scale. I'm going off folks I know at client companies and at scale itself. - idk I've trained a lot of models in my time. It's true that there's an arcane art to training LLMs, but it's wrong that this is somehow unlearnable. If I can do it out of undergrad with no prior training and 3 months of slamming my head into a wall, so can others. (Large LLMs are imo not that much different from small…
Dude you went to Columbia, you probably dont think people that went to state schools are even human. Rest of the article was good
> you probably dont think people that went to state schools are even human
on the contrary. I have been quite vocal about why I felt my education was lacking and the respect I have for those who have gone for nontraditional paths
Re: Meta Spends $14B to Hire a Single Guy
#58I guess OpenAI got a good deal at 6.5b for Jony Ive. >Supposedly, Llama 4 did perform well on benchmarks, but the user experience was so bad that people have accused Meta of cooking the books. This is one of those things that I've noticed. I don't understand the benchmarks, and my usage certainly isn't going to be wide ranging as benchmarks, but I hear "OMG this AI and benchmarks" and then I go use it and it's not an…
OpenAI didn’t get Jony Ive though. They bought an existing joint venture, i.e. he cashed out. They have a contract with his design firm for him to keep working on the project, but it’s basically great for him (massive payout, ongoing payments) and not particularly good for OpenAI.
Re: Meta Spends $14B to Hire a Single Guy
#59Earlier quoted context omitted.
>They have no authority to review a minority investment. That's just wrong. Partial acquisitions and minority shareholdings don't allow you to bypass antitrust investigations. See 15 U.S.C. §18 for example. It is similar in the EU.
You are literally correct, but not engaging in the point. Meta shares in Scale are explicitly non-voting shares. Influence/control post minority investment matters as to whether the deal can be reviewed via Clayton Act. Meta did everything they could to avoid regulatory review with this transaction, and it worked.
The point remains wrong even if you pivot to this argument. Every major competitor has cancelled their deals with Scale AI and there are even rumours that the company will wind down and perhaps eventually stop working while key workers are essentially being absorbed into Meta. This is textbook anti-competitive behaviour and more than enough reason to warrant antitrust investigations by oversight agencies in the major markets.
Re: Meta Spends $14B to Hire a Single Guy
#60Obviously false, and now that the author knows it
> UPDATE: ... much of the $14b did not go to Alexandr
why not change the title as well?