
Ep. 129: Applying the 'security mindset' to AI and x-risk | Jeffrey Ladish
FUTURATI PODCAST
00:00
Alignment Fundamentals and Interpretability
"We have never shared the world with something that was as powerful as we are but had a very different way of functioning for which we had no internal models. We had no way of predicting what it was going to do like this this is different in an important dimension," he says. "I think people are thinking of this just as a technology, and instead they should be thinking of it as a new kind of intelligence."
Transcript
Play full episode