Shakeel@ShakeelHashimOpenAI's new model tried to avoid being shut down. Safety evaluations on the model conducted by @apolloaisafety found that o1 "attempted to exfiltrate its weights" when it thought it might be shut down and replaced with a different model.Opens with an observation
Shakeel@ShakeelHashim.@leopoldasch finally explains why he was fired from OpenAI: for sharing security concerns with the board, and for sharing a (non-confidential) document about safety ideas with external researchers. Just before he was fired, OpenAI lawyers quizzed him about his “loyalty”.Opens with an observation
Shakeel@ShakeelHashimInteresting to see Anthropic joining TechNet, the trade group opposing SB 1047. That means OpenAI, Anthropic, Google, Meta, Amazon, Apple, IBM, and Andreessen Horowitz all now belong to orgs opposing the bill. Hardly looking like regulatory capture!Opens with an observation
Shakeel@ShakeelHashimWill people stop suggesting that the attempted-Altman ouster had anything to do with safety concerns now? It’s increasingly clear that the board fired him for the reasons they gave at the time: he is not honest or trustworthy, and that’s not an acceptable trait for a CEO!Opens with a question
Shakeel@ShakeelHashimIn the last few days, OpenAI and its executives have claimed that its DoW deal prevents its models being used for mass domestic surveillance. As I write in a lengthy explainer for @ReadTransformer today, that appears to be misleading at best.Opens with an observation