They’re planning to update it to use on-device models, their “private compute” models, and optionally for certain prompts (with explicit user approval) OpenAI m

They’re planning to update it to use on-device models, their “private compute” models, and optionally for certain prompts (with explicit user approval) OpenAI models. This is not directly in contradiction to what you said, but I feel like there’s a lot of confusion around this.