Why has nobody trained a good model capable of saying "I don't know"? Is it that hard or is it just that nobody cares?
Siméon
@Simeon_Cps
Building world-models for verified AI inference in London | former founder & CEO of SaferAI
Joined May 2020
- The true state of AI development is so insane that if you tell it to someone outside tech without quoting authorities they think you're lying.
- While you, humans, try and fail to play 4d chess, Altman is casually playing in 6d.
- The fact that in 10 years the world will probably be entirely transformed by AI and that no one is paying attention right now seems ridiculous to me.
- Indeed. We will all be Lee Sedol.
We will all be Lee Sedol - Senators attending Altman’s testimony before congress, to get a clearer picture of the situation in AI, consider asking the 8 following questions: 1) When do you think we'll develop the first AI system that surpasses humans in all cognitive capabilities? (This is referred as
- 40 seconds to understand why the human species might go extinct and why if we race towards AGI, everyone will lose, by @geoffreyhinton
00:00 - Wait, can we consider seriously the hypothesis that 1) the recent hyped tweets from OA's staff 2) "AGI has been achieved internally" 3) sama's comments on the qualification of slow or fast takeoff hinging on the date you count from 4) sama's comments on 10000x researchers are
Replying to @samadoes that mean: "if you think of ChatGPT as the start of the takeoff, then there will indeed be >1y or more btwn it and AGI but if you think it hasn't started yet, then we're in a fast-takeoff"? - It seems like Altman was leveraging his OpenAI streetcred to build new personal AI companies (a compute one and a device one), which: 1. would provide him with potentially enormous amounts of bargaining power with OpenAI, especially the compute one. 2. would make him much more
