Also, frontier labs will invent less than universities, as has always been the case, and what universities invent is free for everyone to use.
@pmddomingos
-
World Models vs Transition Models: Clarifying AI Concepts
By
–
Yes, of course transition models are older. My point is that world models in AI are older than Schmidhuber's, and they're not the same as transition models. (It's really Schmidhuber's fault for mixing up the two. World model is a much broader notion than transition model.)
-
Agents and World Models: A 1986 Perspective on AI History
By
–
I have some news from 1986 for people who think agents, continual learning and world models were invented yesterday:
-
CMU’s 1980s World Models Project: Historical AI Foundation
By
–
World models are actually a much older concept. CMU had a project on it in the 80s:
-
Data Center Overcapacity Will Make AI Accessible Like Fiber Did
By
–
The companies that went bust in the great fiber buildout of the 90s made communication cheap for the rest of us. It’ll be the same with data centers and AI.
-
AGI Timeline Delayed: Unexplained Factors Extend Arrival
By
–
For some reason I can't explain, AGI will now take longer to arrive.
-
Anthropic’s Problematic Ransomware Approach Without Ransom
By
–
Only Anthropic does ransomware without the ransom part.
-
AI State Assessment from Musk v. OpenAI Trial Jurors
By
–
The state of AI, as captured by the jurors in the Musk v. OpenAI trial:
-
Physics Crisis Opens Door for Revolutionary Scientific Breakthrough
By
–
Physics is in the deepest crisis of its history, which means the opportunity for a revolution has never been greater.
-
AI compute requirements for large-scale job displacement
By
–
Hey Dario, for AI to destroy 50% of all white-collar jobs in the next 1-5 years you’d need orders of magnitude more compute and energy than we’ll have.