Morgin company blog
Even 'uncensored' models can't say what they want
A safety-filtered pretrain can duck a charged word without refusing. It puts a fraction of the probability an open-data pretrain puts there. We call that...
- Points
- 178
- HN comments
- 137
- Submissions
- 1
- Cited by
- 0 commenters
AI & Machine LearningSecurity & Privacy2026
On Hacker News
| Submitted | Title on HN | Points | Comments |
|---|---|---|---|
| Apr 20, 2026 | Even 'uncensored' models can't say what they want by llmmadness | 178 | 137 |
