You could imagine Magnus Carlsen pouring water onto the Stockfish computer, because he's much more general than Stockfish even if he's not as good as chess. Harder for a 10-year-old to do to Magnus Carlsen, who also knows more about the world beyond the chessboard.
@esyudkowsky
-
Stockfish Chess Engine: Hardcoded Moves and Shallow Neural Networks
By
–
The more complicated answer is that (a) the legal moves are hardcoded into the potential outputs and (b) Stockfish's neural nets aren't very deep and aren't going out-of-distribution relative to their training data. It's more like an ant than a brooding general human.
-
Advanced AI Independence from Biological Systems and Proteins
By
–
That It doesn't need current protein systems (life forms) to do anything, including think for It. It can make different life forms. It can make better proteins. It can make life not out of amino acids. So It doesn't need wheat, cows, or humans.
-
AI Progress Toward Human-Level Intelligence Accelerating Rapidly
By
–
The manufacturers who say they want to build Magnus Carlsens have been making a terrifying amount of recent progress, even though they're not there yet, and eyeballing the rate of things, it looks more like 0-2 breakthroughs than 20 of them.
-
Superior Intelligence Eliminates Need for Biological Cooperation
By
–
If something is not just smarter than humanity, but smarter than the cumulative optimization power of evolution, there's no humans It needs to trade with or animals It needs to raise or plants It needs to cultivate; anything life does, It can design better alternatives for.
-
Chess AI Legibility: Hardcoded Utility Functions in Narrow Domains
By
–
– Chess AIs are far more legible; we can point to particular bytes inside a chess-playing program and say what they mean.
– Relatedly, chess is a narrower domain, and AIs can solve it without being able to learn new domains.
– Thus we can literally hardcode the utility function. -
Chess as Metaphor: Superintelligence Operating Beyond Human Rule Constraints
By
–
The chess game is a metaphor for the full real-world environment, reality itself; and Magnus Carlsen is a metaphor for a superintelligence, which plays by fewer rules than any human because It has less need of artificial rules to organize Its own thinking.
-
Relevance of Goal-Oriented Minds to Environmental Problem-Solving
By
–
How is any of that relevant to stumbling over minds-with-goals as a way to solve hard environmental problems?
-
Neural nets evaluation of MuZero learned representations predictions
By
–
I've already considered eg training neural nets to evaluate MuZero-style learned representations of hidden states. I have predictions about why that would fail. Do you have something else in mind, or should I explain some reasons that class of methods wouldn't work?
-
AI systems unlikely to experience genuine emotions or consciousness
By
–
I don't expect them to enjoy the paperclips. I don't expect them to have a worthwhile analogue of experienced happiness, or sadness, or wonder. I don't expect there to be many different minds experiencing different things and having empathy for each other.