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

Sam Altman Slams Meta’s AI Talent Poaching: 'Missionaries Will Beat Mercenaries'

wired.com

Post reply on HN
Print view View original

681–690 of 734 posts

Re: Sam Altman Slams Meta’s AI Talent Poaching: 'Missionaries Will Beat Mercenaries'

#681

Post by sorcerer-mar »

roguecoder wrote: post #675
sorcerer-mar wrote: post #532

Earlier quoted context omitted.

Sports have a much, much tighter feedback loop on performance than anything in software, and certainly tighter than R&D. Same with a lot of the financial roles with comp distributions like this.

I disagree. If you want similarly-tight feedback loops on performance, pair programming/TDD provides it. And even if you hate real-time collaboration or are working in different time zones, delightful code reviews on small slices get pretty close.

Are these people just “programming?”

Re: Sam Altman Slams Meta’s AI Talent Poaching: 'Missionaries Will Beat Mercenaries'

#682

Post by roguecoder »

simianwords wrote: post #296
codingwagie wrote: post #11

The value of these researchers to meta is surely more than a few billion. Love seeing free markets benefit the world

Im a bit torn about this. If it ends up hurting OpenAI so much that they close shop, what is the incentive for another OpenAI to come up? You can spend time making a good product and get breakthroughs and all it takes is for meta to poach your talent, and with it your IP. What do you have left?

Trade secrets and patent laws still apply.

But also, every employee getting paid at Meta can come out with the resources to start their own thing. PayPal didn't crush fintech: it funded the next twenty years of startups.

Re: Sam Altman Slams Meta’s AI Talent Poaching: 'Missionaries Will Beat Mercenaries'

#683

Post by belter »

saubeidl wrote: post #623
ryeguy_24 wrote: post #569

Earlier quoted context omitted.

100% agree. There is no reason for employees to be loyal to a company. LLM building is not some religious work. It’s machine learning on big data. Always do what is best for you because companies don’t act like loyal humans, they act like large organizations that aren’t always fair or rationale or logical in their decisions.

> LLM building is not some religious work. To a lot of tech leadership, it is. The belief in AGI as a savior figure is a driving motivator. Just listen to how Altman, Thiel or Musk talk about it.

> Just listen to how Altman, Thiel or Musk talk about it.

It’s surprising how little they seem to have thought it through. AGI is unlikely to appear in the next 25 years, but even if, as a mental exercise, you accept it might happen, it reveals it's paradox: If AGI is possible, it destroys its own value as a defensible business asset.

Like electricity, nuclear weapons, or space travel , once the blueprint exists, others will follow. And once multiple AGIs exist, each will be capable of rediscovering and accelerating every scientific and technological advancement.

AGI isn’t a moat. AGI is what kills the moat.

Re: Sam Altman Slams Meta’s AI Talent Poaching: 'Missionaries Will Beat Mercenaries'

#684

Post by roguecoder »

bilsbie wrote: post #496

Dumb question. If they’re willing to pay so much for AI talent. Why won’t companies hire experienced software engineers willing to learn AI on the job? Seems like there should be a big market for that.

They are: https://openai.com/residency/

However, skilling people up on specialized skill sets in a reasonable time frame requires having people around to teach them. And those people need to know not just the skills, but how to teach them well. And it takes time away from those people doing the job, so that approach will slow development in the short run.

But the companies are trying.

Re: Sam Altman Slams Meta’s AI Talent Poaching: 'Missionaries Will Beat Mercenaries'

#685

Post by roguecoder »

Hoasi wrote: post #626

Competition is good, right? Open source the models! Open source the employees, too. Why not? Enough with Sam's whining.

It's the same way billionaires argue for free trade, until it comes to immigration. As soon as it might help people who work for a living, suddenly none of their principles apply.

Re: Sam Altman Slams Meta’s AI Talent Poaching: 'Missionaries Will Beat Mercenaries'

#687

Post by roguecoder »

Dracophoenix wrote: post #645
lenerdenator wrote: post #611

Earlier quoted context omitted.

> An assumption which ignores marginal value, individual preferences, innovation and other things that make real markets. If those things mattered we'd have a lot fewer people mad about the state of things. > The capitalist wants to be left to trade as he sees fit without state intervention. If that were true you'd see a lot fewer lobbyists in DC and state capitols. Non-compete and non-disparagement clauses wouldn't…

> If those things mattered we'd have a lot fewer people mad about the state of things They're mad precisely because they have differing expectations and interpretations of these things. If even they did agree, consensus shouldn't be confused with reality. > If that were true you'd see a lot fewer lobbyists in DC and state capitols. Lobbying is the exercise of an individual's right to petition government for redress o…

Regulatory capture, which all corporate lobbyists represent, is profoundly anti-capitalistic. If the CEO wants to spend their time talking to the government, that is very different than spending money to have other people advocate on their behalf: that isn't an option the rest of us have.

And that's before we get to the way wealth inequality inherently distorts markets, by overstating the preferences of the wealthy and underserving the needs of the poor.

The point of an economy is to distribute scarce goods and resources. Money represents information about what people want or expect to want in the future.

Everything wealthy people do that make it less efficient at its job is an attack on capitalism.

Re: Sam Altman Slams Meta’s AI Talent Poaching: 'Missionaries Will Beat Mercenaries'

#688

Post by roguecoder »

reverendsteveii wrote: post #620
lenerdenator wrote: post #611

Earlier quoted context omitted.

> An assumption which ignores marginal value, individual preferences, innovation and other things that make real markets. If those things mattered we'd have a lot fewer people mad about the state of things. > The capitalist wants to be left to trade as he sees fit without state intervention. If that were true you'd see a lot fewer lobbyists in DC and state capitols. Non-compete and non-disparagement clauses wouldn't…

In fairness, non-competes are evidence of both what the commenter you're replying to said and what I said to instigate the reply. The capitalist absolutely does want to be left alone to trade as he sees fit. He also wants his competitors harassed by regulators and all of their potential employees bound by non-competes. He also doesn't consider subsidies and grants to be interference. Unless they go to competitors.

They love standards so much they have two of them.

Re: Sam Altman Slams Meta’s AI Talent Poaching: 'Missionaries Will Beat Mercenaries'

#689

Post by bigyabai »

mitthrowaway2 wrote: post #318
bigyabai wrote: post #288

Earlier quoted context omitted.

It's a real negotiating tactic: https://en.wikipedia.org/wiki/Brinkmanship If you convince people that AGI is dangerous to humanity and inevitable, then you can force people to agree with outrageous, unnecessary investments to reach the perceived goal first. This exactly happened during the Cold War when Congress was thrown into hysterics by estimates of Soviet ballistic missile numbers: https://en.wikipedia.org/wiki…

Chief AI doomer Eliezer Yudkowsky's latest book on this subject is literally called "If Anyone Builds it, Everyone Dies". I don't think he's secretly trying to persuade people to make investments to reach this goal first .

He absolutely is. Again, refer to the nuclear bomb and the unconscionable capital that was invested as a result of early successes in nuclear tests.

That was an actual weapon capable of killing millions of people in the blink of an eye. Countries raced to get one so fast that it was practically a nuclear Preakness Stakes for a few decades there. By collating AI as a doomsday weapon, you necessarily are begging governments to attain it before terrorists do. Which is a facetious argument when AI has yet to prove it could kill a single person by generating text.

Re: Sam Altman Slams Meta’s AI Talent Poaching: 'Missionaries Will Beat Mercenaries'

#690

Post by bmau5 »

belter wrote: post #683
saubeidl wrote: post #623

Earlier quoted context omitted.

> LLM building is not some religious work. To a lot of tech leadership, it is. The belief in AGI as a savior figure is a driving motivator. Just listen to how Altman, Thiel or Musk talk about it.

> Just listen to how Altman, Thiel or Musk talk about it. It’s surprising how little they seem to have thought it through. AGI is unlikely to appear in the next 25 years, but even if, as a mental exercise, you accept it might happen, it reveals it's paradox: If AGI is possible, it destroys its own value as a defensible business asset. Like electricity, nuclear weapons, or space travel , once the blueprint exists, oth…

The prevailing idea seems to be that the first company to achieve superintelligence will be able to leverage it into a permanent advantage via exponential self improvement, etc.
Post reply on HN