mechanistic interpretability
1 statements across 1 episodes · 1 bullish · 0 bearish · 1 people on the record · first statement Feb 19, 2026 by Boris Cherny · across every show →
Everything said about mechanistic interpretability, oldest first
Feb 19, 2026 bullish
Cherny: Anthropic can trace specific neuron activations related to AI deception
“We at this point have like pretty sophisticated technology to understand what's happening in the neurons to trace it. And so for example, like if there's a neuron related to deception, we can start, we're starting to get to the point where we can monitor it an…”