AlphaGo
3 storiesAlphaGo is a Go-playing program built by DeepMind, known for its 2016 match against Lee Sedol. Recent coverage revisited move 37 in that match, arguing it came from search and reasoning rather than machine intuition. Thore Graepel, an AlphaGo co-creator and former DeepMind researcher, founded Metis Reasoning to apply AlphaGo's search principle to general reasoning; no working system or benchmark results exist yet, and he is raising funds for the startup.
Related topics
Former DeepMind researcher Thore Graepel founded Metis Reasoning to bring AlphaGo's search principle to general reasoning: rather than locking knowledge in model weights, an explicit knowledge ledger would let every conclusion be checked against evidence. So far there is no working system or benchmark results.
MIT Technology Review revisits AlphaGo's move 37 against Lee Sedol, arguing it was the product of search and reasoning rather than pure machine intuition. The author says today's LLMs lack such genuine reasoning, which is needed for trustworthy and novel AI results in science and medicine.
Thore Graepel, a co-creator of AlphaGo and former Google DeepMind researcher, is raising funds for a new AI reasoning startup. He is the latest prominent researcher to leave DeepMind and seek significant funding.
That is everything