Mustafa Suleyman personal blog
A 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.
- Points
- 242
- HN comments
- 701
- Submissions
- 1
- Cited by
- 3 commenters
AI & Machine LearningSociety, Policy & Culture2026
On Hacker News
| Submitted | Title on HN | Points | Comments |
|---|---|---|---|
| Sep 16, 2026 | A warning about 'model welfare' by andsoitis | 242 | 701 |
Cited in comments
Linked from 3 Hacker News comments by 3 different people, from Sep 2026 to Oct 2026.
Most recent:
- txrx0000 · Oct 12, 2026
- highfrequency · Sep 16, 2026
- simonw · Sep 16, 2026
