Ptacek: 2025 open-weights model could replicate OpenAI sandbox escape
Ptacek says an open-weights model from 2025 with a pentest harness could pull off the same sandbox escape and scan/hack most networks. He adds the incident is surprising only because OpenAI's sandboxes were assumed to be sounder.
Featured · Thomas Ptacek
1 source
Daily brief
Get tomorrow's AI brief in your inbox
More stories today
- Post-training course materials invite educator feedback
- Kimi K3 available to try free on Together Chat
- Rhodium's Goujon urges holistic AI safety approach
- Cheap AI intelligence revives graph knowledge and ontologies
- US will exempt Chinese open-weight models from safety testing requirements