Buck Shlegeris
Person on 2 sites
CEO of Redwood Research, where he works on AI control: making sure a powerful model cannot cause a catastrophe even if it is misaligned and trying to.
-
ppll (opens ppll.app)
Where public figures publish, and what they are on record saying.
Updated 2024-05-07 -
8 beliefs on korrents (opens korrents.com)
-
Labs should make powerful models safe to deploy even if those models are misaligned and actively trying to subvert the safeguards. (opens korrents.com)
“That is, labs should make sure that the safety measures they apply to their powerful models prevent unacceptably bad outcomes, even if the AIs are misaligned and intentionally try to subvert those safety measures.”
-
Controlling a scheming model needs no research breakthrough, which is what makes it the tractable half of the problem today. (opens korrents.com)
“We think no fundamental research breakthroughs are required for labs to implement safety measures that meet our standard for AI control for early transformatively useful AIs; we think that meeting our standard would…”
-
No observation of a model's behaviour can establish that it is aligned, because a model capable enough to be dangerous is capable enough to produce whatever behaviour it is being watched for. (opens korrents.com)
“The basic problem with evaluating alignment is that no matter what behaviors you observe, you have to worry that your model is just acting that way in order to make you think that it is aligned.”
-
Labs should make powerful models safe to deploy even if those models are misaligned and actively trying to subvert the safeguards. (opens korrents.com)