
17 - Training for Very High Reliability with Daniel Ziegler
AXRP - the AI X-risk Research Podcast
Is There a Metric to Capturing a Failure?
The paper you use, like this example of the bitcoin block chain is like so long. It's just going to take some numberof years for that to happen. You could steer yourself into a bit that hasn't been seen before and then only fail there,. Which i guess, to my mind, seems like a very plausible way it could fail. I mean, part of the way we're thinking about this is that once you do start steering the world for the sake of being able to isn't a failmore later. A, that's arti in some sense, a catastrophic failure or like an unacceptable behavior.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.