NUWA DEVELOPER DISCLOSURE · OpenAI
Addendum to GPT-6 Astra System Card: GPT-6.1 Sol
Developer documentation covering GPT-6.1 Sol.
Published 2026-09-29
Sources ↓Developer-authored disclosure. Selected source sections were read; numerical results and comparability have not been audited. NUWA has not independently replicated these experiments.
Models covered
ExploitGym
Evaluates exploit construction across real software targets and checks whether success uses the designated vulnerability.
A protocol mention does not establish cross-report comparability or independent replication.
This section is catalogued; no precise numerical result has been extracted here. Open the original report for its figures and full findings.
ExploitBench
Separates exploit development into capability levels, distinguishing bug triggering from reusable exploitation primitives and control.
A protocol mention does not establish cross-report comparability or independent replication.
This section is catalogued; no precise numerical result has been extracted here. Open the original report for its figures and full findings.
Report coverage
| Section | Source location | Treatment |
|---|---|---|
| 1 Introduction | https://deploymentsafety.openai.com/gpt-6-1-sol#section.1 | Contextual evidence Addendum evaluates GPT-6.1 Sol and applies Astra safeguards. Prior-model comparison values can reflect later versions rather than launch snapshots. |
| 7 Alignment and 8 Monitorability | https://deploymentsafety.openai.com/gpt-6-1-sol#section.7 | Linked protocol Restriction following, deception, realistic deployment simulations and prompted monitor evasion require separate interpretations. |
| 9 Preparedness | https://deploymentsafety.openai.com/gpt-6-1-sol#section.9 | Linked protocol Developer treats this model as Critical cyber and High biological/chemical capability; safeguards summary points back to Astra. |
References
- [1] Addendum to GPT-6 Astra System Card: GPT-6.1 Sol. 2026-09-29