EAG 2018 London: Reframing Superintelligence

In the early days of debate on advanced artificial intelligence, we didn’t yet have very good AI systems at all, so we had to rely on ideas about human thinking and extrapolate. Now, however, there are plenty of advanced AI systems, many of which are better than humans at narrow tasks. In this talk from EA Global 2018: London, Eric Drexler argues that using this new data, we should rethink our models for how superintelligent AI is likely to emerge and function. You can read a transcript of this talk here: https://www.effectivealtruism.org/articles/ea-global-2018-reframing-superintelligence/