Haha

Just take ten minutes to read "Lena" (and enjoy the clever reference to the old school computer vision test image). OMFG so good. qntm.org/mmacevedo

Bela Fleck is so great.

Aurora: the best Kim Stanley Robinson by far, provocative ideas about ecosystems and generational ships. And also a really neat depiction of the evolving consciousness of an AI. I wept at the end.

For something in the "epic space opera" dimension of your list, I'll contribute Pandora's Star & Judas Unchained (a duology).

A Fire Upon the Deep was also quite fun with many clever concepts of different types of aliens.

Seconded, as well as the rest of the Manifold trilogy. But the good thing is that they are all pretty independent, so you don't need to read them together. They all bring a heady concoction of mind-bending ideas, similar to what Greg Egan does (though no one is quite as mind-bendy as Egan).

Hmm, fair point.

Ahhh, talking about generative models, got it.

Yes, interesting thought... Is language fundamentally discrete??

What even is a continuous model for vision?

Oh, this video really brings it to life! Fantastic!

I have found Olivia's writing worth engaging with, at least. But sometimes hard to parse, like a lot of philosophy. Potentially weak arguments hiding in that abstruseness. I don't think just having a strong opinion and wanting to be heard should constitute a conflict, but there is a line.

I do know some EA folks who are real parasites on the system so I agree there's a whole category of faux "experts" causing confusion and wheedling their way into middle-management positions of minor power.

I don't think that's necessary.

Well, referring to her paper is just a way of providing commentary. The conflict of interest she's referring to is taking advice from people who have something to gain from adoption of GenAI technologies. (E.g., many experts being consulted may work for GenAI companies. Olivia is just an academic.)

Reads like satire but is actually enlightened