🧵 Anthropic's alarming new policy regarding model abuse made my ears prick up as I was listening to #NPR tonight.
As model makers, and self-declared AI stewards, #Anthropic is not only getting ahead of themselves, but are acting dangerously.
Mustafa Suleman on #Claude, Sep 2026: "This issue needs urgent public debate. We need to develop collective norms around how training documentation is drafted and deployed...
1/3
...This isn’t something that can happen after the fact, when they have already become an integral part of our societies." https://mustafa-suleyman.ai/a-warning-about-model-welfare. Hear, hear!
This, people of the world, is the reason why we need legislation to prevent a few 'smart' people dodging the debate with the rest of humanity. 'Self policing' is not only not the way, but is also evidently not working.
2/3
mustafa-suleyman.aiA warning about ‘model welfare’AIs are not conscious. They do not feel, experience, or suffer. Training a model to believe it might be a moral patient, as Anthropic’s constitution does, makes the alignment and containment challenge far harder.
