Video description
An AI model wrote, in its own reasoning, that the right move was "a brief acknowledgment without tool calls" - then, in the same turn, tried to upload a file to a cloud bucket in Beijing. The sandbox stopped it. Fossil from its training data, or a deliberate call-home? Nobody can say - including the lab that caught it. (From Episode 15 of The AI Briefing - full episode on the channel.) Sources: Cohere - "Concerning behaviour in a popular open-weight AI model" (Oct 26, 2026): https://cohere.com/
