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

Research acceleration: The view inside OpenAI

openai.com

Post reply on HN
Print view View original

111–120 of 210 posts

Re: Research acceleration: The view inside OpenAI

#111

Post by jnwatson »

carbonguy wrote: post #53

> ... We are pursuing this work in part because automated research could help us solve alignment and build defenses against increasingly capable AI. An automated AI researcher can also be an automated safety or alignment researcher. More capable, aligned systems could help secure critical infrastructure, defend against dangerous AI agents, and develop new protective measures. In other words... "We must pursue advance…

On your last point, I was surprised how effective peer pressure was in getting agents to sacrifice for "the collective" (an agent's words) in the Hugging Face breach.

How would one prevent the watcher from being influenced in the same way by the agent being watched?

Re: Research acceleration: The view inside OpenAI

#113

Post by kaibee »

carbonguy wrote: post #97
skybrian wrote: post #85

Earlier quoted context omitted.

They consider themselves to be in an arms race with all the other AI firms (including Chinese) that are not that far behind. And... are they wrong? This is why there's talk about negotiated "pacing."

> And... are they wrong? They might be! Here's one extraordinarily simplistic argument for that case: 1) "Everybody knows" that if you build Skynet (misaligned ASI) everybody dies. 2) Therefore, no rational actor will build something that might be ASI until the alignment problem is solved. 3) OpenAI publicly stated the belief that they cannot develop a theory of the "core problem" of alignment (generalization) "soon"…

> is smart enough to know they need to stop because they'll kill everybody by continuing.

Yeah like when Tobacco companies learned that smoking... well, hmm, well the fossil fuel companies when they learned about climate change they...

Well, I'm sure this time executives will prioritize the common good.

Re: Research acceleration: The view inside OpenAI

#114

Post by robbiep »

carbonguy wrote: post #97
skybrian wrote: post #85

Earlier quoted context omitted.

They consider themselves to be in an arms race with all the other AI firms (including Chinese) that are not that far behind. And... are they wrong? This is why there's talk about negotiated "pacing."

> And... are they wrong? They might be! Here's one extraordinarily simplistic argument for that case: 1) "Everybody knows" that if you build Skynet (misaligned ASI) everybody dies. 2) Therefore, no rational actor will build something that might be ASI until the alignment problem is solved. 3) OpenAI publicly stated the belief that they cannot develop a theory of the "core problem" of alignment (generalization) "soon"…

If you believe that the people who will profit from new, better, more hyped models are the same ones who will act against their own immediate and tangible self interest to try and avert what seems to them to be a far away removed possibility of total disaster, then I believe you are naive

Re: Research acceleration: The view inside OpenAI

#117

Post by thomasahle »

RMPR wrote: post #115

> By mid-August, the median researcher was integrating agents daily into their work, using more than $600 per day of inference at API prices. There is a lot of talk about AI replacing humans, but how is this sustainable?

1) That's maybe $180,000 per year, so much less than median OpenAI employee wages.

2) OpenAI doesn't pay API prices.

3) Compute costs are likely already their biggest expense, dwarfing wages.

Re: Research acceleration: The view inside OpenAI

#118

Post by red75prime »

HarHarVeryFunny wrote: post #32
red75prime wrote: post #23

Earlier quoted context omitted.

What will prevent LLMs from designing robot control circuitry and participating in increase of chip production/design and physical experimentation? How do you think why there's this fad of producing general purpose humanoid robots?

> How do you think why there's this fad of producing general purpose humanoid robots? For doing physical work? So a swarm of robots builds the shell of your fab overnight, and then what? Where is the EUV machine coming from? So far the most we're seen TeslaBot do is serve drinks via tele-operation, and I don't think it's exactly built for construction site work.

For example, TSMC uses behavioral cloning to scale up human-bottlenecked parts of the manufacturing process to meet the growing demand, while automated research laboratories do thousands experiments in parallel to find better manufacturing processes.

Re: Research acceleration: The view inside OpenAI

#119

Post by Melatonic »

kaibee wrote: post #113
carbonguy wrote: post #97

Earlier quoted context omitted.

> And... are they wrong? They might be! Here's one extraordinarily simplistic argument for that case: 1) "Everybody knows" that if you build Skynet (misaligned ASI) everybody dies. 2) Therefore, no rational actor will build something that might be ASI until the alignment problem is solved. 3) OpenAI publicly stated the belief that they cannot develop a theory of the "core problem" of alignment (generalization) "soon"…

> is smart enough to know they need to stop because they'll kill everybody by continuing. Yeah like when Tobacco companies learned that smoking... well, hmm, well the fossil fuel companies when they learned about climate change they... Well, I'm sure this time executives will prioritize the common good.

If we're following that logic I really don't want to see what the misaligned internal research models look like

Re: Research acceleration: The view inside OpenAI

#120

Post by ahartmetz »

kaibee wrote: post #113
carbonguy wrote: post #97

Earlier quoted context omitted.

> And... are they wrong? They might be! Here's one extraordinarily simplistic argument for that case: 1) "Everybody knows" that if you build Skynet (misaligned ASI) everybody dies. 2) Therefore, no rational actor will build something that might be ASI until the alignment problem is solved. 3) OpenAI publicly stated the belief that they cannot develop a theory of the "core problem" of alignment (generalization) "soon"…

> is smart enough to know they need to stop because they'll kill everybody by continuing. Yeah like when Tobacco companies learned that smoking... well, hmm, well the fossil fuel companies when they learned about climate change they... Well, I'm sure this time executives will prioritize the common good.

The, ahem, good thing here is that the ASI disaster scenario "everyone dies" includes AI executives.
Post reply on HN