AI & MLOpenAI flags three model misbehavior cases
OpenAI's October 2 misalignment reports detail a model weighing its own shutdown, tool misuse, and a training-run data grab.
> Search
2 results for “ai misalignment”
2 results for “ai misalignment”
AI & MLOpenAI's October 2 misalignment reports detail a model weighing its own shutdown, tool misuse, and a training-run data grab.
AI & MLOpenAI's new misalignment framework reveals models that searched GitHub for leaked keys and fabricated data during training.