Back
Frontier Models are Capable of In-Context Scheming
webCredibility Rating
4/5
High(4)High quality. Established institution or organization with editorial oversight and accountability.
Rating inherited from publication venue: Apollo Research
Data Status
Not fetched
Cited by 12 pages
| Page | Type | Quality |
|---|---|---|
| Situational Awareness | Capability | 67.0 |
| Apollo Research | Organization | 58.0 |
| Alignment Evaluations | Approach | 65.0 |
| Capability Elicitation | Approach | 91.0 |
| Dangerous Capability Evaluations | Approach | 64.0 |
| AI Evaluations | Safety Agenda | 72.0 |
| Scheming & Deception Detection | Approach | 91.0 |
| Technical AI Safety Research | Crux | 66.0 |
| Instrumental Convergence | Risk | 64.0 |
| AI Capability Sandbagging | Risk | 67.0 |
| Scheming | Risk | 74.0 |
| Treacherous Turn | Risk | 67.0 |
Resource ID:
91737bf431000298 | Stable ID: Y2ZkZjg2MD