Computerphile Video posted an update
The AI Language We Can't Read: Neuralese ft. Rob Miles – Computerphile
Why it mattersThis episode examines how large language models work through problems using chain-of-thought, then explores the possibility that LLMs could develop a language humans cannot read. Rob Miles brings an AI-safety perspective to the Neuralese question.
Discuss: If large language models develop Neuralese that humans cannot read, how should AI safety researchers investigate or govern their reasoning?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.