Jeffrey Ladish

speaker

1,006 appearances 1 recordings 1 series first heard Apr 2025 last heard Apr 2025

Jeffrey Ladish’s voice in public audio — every appearance, attributed to the second.

Trend

recordings per month · last 12 months

No recordings in the last 12 months.Older appearances are listed below; set an alert to hear about the next one.

Appearances

newest first · ▶ plays the moment

"The Cognitive Revolution" · Reward Hacking by Reasoning Models & Loss of Control Scenarios w/ Jeffrey Ladish of Palisade Research, from FLI Podcast · 2 Apr 2025 podcast

the place like the you know, these the systems will have goals because we'll be

training them on on tasks that require them to have goals.

Um or at least to have goal directed behavior.

I'm not making a claim about like what it feels like to be the AI.

But I'm saying when we look at its behavior, it's going to behave in strongly goal directed ways.

Like in the in the O one example with with hacking, you know, it's gonna be like I wanna solve this thing, so I'm gonna find a creative solution to solve it.

But I don't I think the the where where the goals come from are sort of things that worked well

in their training environment.

And these can often be pretty simple, but not necessarily things that we we like or things that we want.

And we just have very little way of distinguishing between

Hey, the AI systems like, you know, are are doing things because we want them to, versus the AI systems are doing things because they're smart enough to realize that y that while we have control over them, they need to act in ways that are aligned with us.

Um and so I I expect that, you know, even if AI systems like look pretty aligned, it's really hard to tell whether they actually are.

And it's just not safe to train relentless problem solvers.

And also try to you know and and also hope that, you know, they'll be really nice to us when they have have more power than us.

Yeah, so this is an interesting question.

I think it's it's it kind of depends on who the defenders and who the attackers are.

So and I wanna caveat all this by saying, while we can control them, like I think like the ultimate winners in sort of the offense-defense balance will be the AI systems themselves.

And I think this is uh so once they are better than humans across the board, they'll be both better at defense and better at offense than us.

But in the interim, while we still like you know, while we so they're not very strategic and we can still mostly control them, I think that

One question is access.

Showing 601–620 of 1,006 · page 31 of 51