while it’s easy to agree with some abstract version of “upgrade” (as in try to channel AI capability gains into our ability to align them), the main bottleneck to physical upgrading is the speed difference between silicon and wet carbon: https://www.lesswrong.com/posts/Ccsx339LE9Jhoii9K/slow-motion-videos-as-ai-risk-intuition-pumps
Yeah to be clear I don’t think “upgrading” is easy. It might not even be possible in a way that makes it relevant. But I do think it offers some hope in an otherwise pretty bleak landscape.
while it’s easy to agree with some abstract version of “upgrade” (as in try to channel AI capability gains into our ability to align them), the main bottleneck to physical upgrading is the speed difference between silicon and wet carbon: https://www.lesswrong.com/posts/Ccsx339LE9Jhoii9K/slow-motion-videos-as-ai-risk-intuition-pumps
Yeah to be clear I don’t think “upgrading” is easy. It might not even be possible in a way that makes it relevant. But I do think it offers some hope in an otherwise pretty bleak landscape.