Lex Fridman
3 HN recommenders · recommended in 5 months, 2020–2022
1:27:57 · 2.7M views · Mar 31, 2020
Lex Fridman interviews physicist Roger Penrose on consciousness, the limits of computation, quantum mechanics and his cyclic cosmology. It covers his views on AI and the big bang.
HN recommends it as an accessible walkthrough of Penrose's ideas on consciousness and cosmology, interesting even if debatable.
6 comments from Hacker News
Lex Fridman interviewed Penrose on this topic recently (2020) and Penrose delivered a very accessible walkthrough of his historical and current thoughts on the topic. this video (Deep-link half-way through to his discussion on microtubules)
Such a joy to see Sir Roger Penrose awarded with this honor. One of my favorite contemporary intellectuals. If anyone hasn't seen it, I recommend his recent appearance on the Lex Fridman podcast: this video
Consciousness is Not a Computation (Roger Penrose) | AI Podcast Clips [LINK] Full episode: this video I don't take Roger's view about consciousness for granted but it is very interesting nevertheless.
He doesn't say it's a new universe, e.g. in an interview with Lex Fridman[1] he calls that out and says it explicitly isn't a separate parallel universe. It's more an argument that if you don't have mass because all the mass has decayed, through Einstein's relation of mass and energy and Planck's relation of energy and frequency, and frequency involving distance and time, with no mass, you lose notions of distance and time. Then the infinitly large distant future universe and the infinitely small distant past big bang become indistinguishable with the loss of "large" and "small" to distinguish them. They are both "at infinity", they both "contain" all the energy in the universe, and what we see as a tiny big bang, aliens on the other side saw as a huge expanded universe. From what I (layperson) can tell, it's like the expand/crunch cyclic universe theory, but without the need to crunch back down again. [1] this video
It doesn't get moved so much as we discover that a certain mathematical manipulation plus all our previous mathematical manipulations don't amount to an AI. It also seems like with every breakthrough this complaint is raised without addressing the underlying issues with the new AIs shortcomings. My hot take: true AI is so far out of reach of our ability it's not even funny, which makes the whole field either a search for the fountain of youth at worst, or at best a search for tools to inform humans or to replace humans in rote tasks. See this video