Your Model is Wrong
Why does AI’s confident wrongness feel so offensive? Its model of the world is wrong, but so is our model of it. I think we need to build a new Theory of Mind for AI.
Field notes on AI systems, platform engineering, interpretability, safety, and the occasional philosophical detour.
Why does AI’s confident wrongness feel so offensive? Its model of the world is wrong, but so is our model of it. I think we need to build a new Theory of Mind for AI.