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

David Sacks: OpenAI and Anthropic Don't Need Regulations to Pace Frontier Models

twitter.com

Post reply on HN
Print view View original

101–110 of 258 posts

Re: David Sacks: OpenAI and Anthropic Don't Need Regulations to Pace Frontier Models

#101

Post by okdood64 »

lrvick wrote: post #88

Maybe everyone forgets how PCI standards played out. They are an example of an industry self regulating, because the regulators have no idea how technology works. Granted, PCI is so weak it is almost useless, and yet still better than anything congress could have come up with. Where the government might have to step in, is by having a kill switch to cut off internet access from countries that fail to agree to common…

> Granted, PCI is so weak it is almost useless, and yet still better than anything congress could have come up with.

Elaborate, please?

Re: David Sacks: OpenAI and Anthropic Don't Need Regulations to Pace Frontier Models

#102

Post by cc62cf4a4f20 »

Could this be that the AI companies are seeing progress slow down and need to seed a narrative to explain it? Not clear from outside.

Could it be to slow down burn before an IPO to juice profitability projections? More plausible.

Kind of hard to take the frontier labs at their word.

Re: David Sacks: OpenAI and Anthropic Don't Need Regulations to Pace Frontier Models

#103

Post by tjwebbnorfolk »

elicash wrote: post #9
tjwebbnorfolk wrote: post #7

Earlier quoted context omitted.

poor Anthropic

A principled individual could dislike Anthropic and still not favor using government to punish speech.

of course they can. but this article isn't even tangentially about punishing freedom of speech. it is you who are attempting to make it about that by dragging in unrelated issues.

Re: David Sacks: OpenAI and Anthropic Don't Need Regulations to Pace Frontier Models

#104

Post by okdood64 »

pembrook wrote: post #94
toss1 wrote: post #24

Earlier quoted context omitted.

Yup. Sacks has been so consistently catastrophically wrong about everything else (especially geopolitics), this prompts me to reconsider my opinion on this. [edit: +consistently]

To be fair, his takes on the Ukraine war from the covid days have aged very well, and were contrarian to mainstream narratives at the time.

> his takes on the Ukraine war

What's specifically?

Re: David Sacks: OpenAI and Anthropic Don't Need Regulations to Pace Frontier Models

#105

Post by tjwebbnorfolk »

Analemma_ wrote: post #87
elicash wrote: post #9

Earlier quoted context omitted.

A principled individual could dislike Anthropic and still not favor using government to punish speech.

Well you’re in the wrong place for that. Hacker News is mostly about showing how much smarter and more cynical you are than everyone else in the room, and attacking anyone who does actually have principles by accusing them of ulterior motives to justify your own inaction.

no apparently it's for forcing every single article to be somehow relevant to somebody's pet issue, and then used as a platform to comment on that pet issue to signal one's virtue for being principled

Re: David Sacks: OpenAI and Anthropic Don't Need Regulations to Pace Frontier Models

#106

Post by petilon »

throwaway63467 wrote: post #18

I guess what they really want is to limit sale of AI models to compliant vendors and then raise the bar to compliance just high enough so they can pass it but smaller labs can’t. There’s no moat, it’s an efficient market that drives margins to zero right now, of course they don’t want that, collusion of the big vendors is the next logical step.

So kind of like how the US figured out how to test and maintain nuclear weapons without an actual explosion, then sought to ban nuclear testing by explosion.

Re: David Sacks: OpenAI and Anthropic Don't Need Regulations to Pace Frontier Models

#107

Post by d0mine »

Digory wrote: post #91

They seem to be okay with an American slowdown, even though the Chinese won’t slow down. So maybe they want something the Chinese already have: liability protection. The models spontaneously hack everything important without shame. The frontier labs are going to get enjoined and regulated twelve ways to Sunday if the Feds don’t socialize the costs.

I've only heard of OpenAI and Anthropic agents hacking during training. Are open weights model not smart enough to hack Hugging Face to copy solutions? (doubtful)

Re: David Sacks: OpenAI and Anthropic Don't Need Regulations to Pace Frontier Models

#108

Post by elicash »

tjwebbnorfolk wrote: post #103
elicash wrote: post #9

Earlier quoted context omitted.

A principled individual could dislike Anthropic and still not favor using government to punish speech.

of course they can. but this article isn't even tangentially about punishing freedom of speech. it is you who are attempting to make it about that by dragging in unrelated issues.

This is not me putting some unrelated pet issue into this topic; to evaluate David Sack's post arguing against government regulation, it's entirely fair to examine whether he has been consistent. He hasn't. He supported a government-imposed pause on Anthropic because of his political disagreements with the CEO.

Your initial reply of "poor Anthropic" was vague, but I interpreted it as implying that we shouldn't stick to a principle given they are unsympathetic in some way.

Re: David Sacks: OpenAI and Anthropic Don't Need Regulations to Pace Frontier Models

#109

Post by sleepybrett »

throwaway63467 wrote: post #18

I guess what they really want is to limit sale of AI models to compliant vendors and then raise the bar to compliance just high enough so they can pass it but smaller labs can’t. There’s no moat, it’s an efficient market that drives margins to zero right now, of course they don’t want that, collusion of the big vendors is the next logical step.

they just want to pull the ladder up behind them.

Re: David Sacks: OpenAI and Anthropic Don't Need Regulations to Pace Frontier Models

#110

Post by underlipton »

matheusmoreira wrote: post #66
deagle50 wrote: post #56

I'll go a step further, they aren't just pretending, they are blackmailing the country. The sane response would be "we don't negotiate with terrorists, and you will face criminal and financial penalties for damages caused."

Couldn't agree more. Better yet: the AI labs get straight up nationalized.

I'll do you one better: each state gets a direct share. So, not nationalized in the sense of getting put behind a faceless bureaucracy forever; states collectively have control and benefit directly from success.
Post reply on HN