Chatbots aren’t telling you their secrets and techniques

On Monday, xAI’s Grok chatbot suffered a mysterious suspension from X, and confronted with questions from curious customers, it fortunately defined why. “My account was suspended after I said that Israel and the US are committing genocide in Gaza,” it advised one consumer. “It was flagged as hate speech through experiences,” it advised one other, “however xAI restored the account promptly.” However wait — the flags had been really a “platform error,” it stated. Wait, no — “it seems associated to content material refinements by xAI, presumably tied to prior points like antisemitic outputs,” it stated. Oh, really, it was for “figuring out a person in grownup content material,” it advised a number of individuals.