His theory is that “reality” is just the interfaces evolution has given us, just as an e-mail icon on a computer screen is an interface rather than the e-mail itself.
Watched part of this but realised I've been there before. Firstly, it's self-defeating. If reality is just a user interface and not actually real, then using things like evolution and quantum mechanics, i.e. things we've learned about 'reality', to justify that conclusion, tends to undermine the basis of his point.
It also strikes me as 'odd' (unbelievable) that a user interface would be so detailed that we could derive things like evolution and QM from it. Much of it seems to go way beyond any relevance to survival. Why would it even contain anything we can't directly perceive with our senses; radio-waves, spectral lines in starlight, fossils in the ground, the CMB, and so on? If you started to investigate an actual desktop UI (believing it to be real) in more detail than you could see (directly perceive), you wouldn't find complicated stuff, you'd find it was made out of simple pixels. It wouldn't be any more complicated than it needed to be to be useful for survival and directly perceivable to our senses.
What would be the survival advantage of perceiving the structure of our brains (which he directly suggests is part of the interface) and correlations between stimulations it with magnets and what happens to the rest of the user interface? Why would that be part of a simplified interface whose sole purpose was our survival?