Dario Amodei: evaluation

Transcript context

…How difficult is the second, the latter? Yeah. I think for ASL-3, it’s primarily about security and about filters on the model relating to a very narrow set of areas when we deploy the model. Because at ASL-3, the model isn’t autonomous yet. And so you don’t have to worry about the model itself behaving in a bad way even when it’s deployed internally. So I think the ASL- 3 measures are, I won’t say straightforward, they’re rigorous, but they’re easier to reason about. I think once we get to ASL-4, we start to have worries about the models being smart enough that they might sandbag tests, they might not tell the truth about tests. We had some results came out about sleeper agents and there was a more recent paper about, “Can the models mislead attempts to sandbag their own abilities, present themselves as being less capable than they are?” And so I think with ASL-4, there’s going to be an important component of using other things than just interacting with the models. For example, interpretability or hidden chains of thought where you have to look inside the model and verify via some other mechanism that is not as easily corrupted as what the model says, that the model indeed has some property. So we’re still working on ASL-4. One of the properties of the RSP is that we don’t specify ASL-4 until we’ve hit ASL-3. And I think that’s proven to be a wise decision because even with ASL-3, again, it’s hard to know this stuff in detail, and we want to take as much time as we can possibly take to get these things right. So for ASL-3, the bad actor will be the humans.…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.