Generally, this seems an obvious and correct decision for Tor. Barring integration with a locally run LLM, AI doesn't make sense for the Tor security posture -

Generally, this seems an obvious and correct decision for Tor. Barring integration with a locally run LLM, AI doesn't make sense for the Tor security posture - you don't want to be routing content to unintended/insecure third parties, period.