Indicators / series
evidence.safety_brake_events.for.pt
Source: OpenAI research posts (OpenAI) · OpenAI; short quotation · CSV
OpenAI
| id | text | grade | as of date | published date | tier | audited vs reported | extraction method | review status | retrieved at | http status | content hash | supersedes id | url | flags |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 67d75927e7d23e32 | Astra internal activities paused pending strengthened security controls (7 Aug 2026) | grade D, evidence tier 7 | 2026-08-07 | 2026-08-07 | 7 | company_stated | manual | approved | 2026-09-10T06:28:36.217659Z | 200 | 5f01e41ce61f | — | source page | |
| 82685d1efcfba981 | two-week pause in RL training on deployment-bound models after the Hugging Face incident (Jul-Aug 2026) | grade D, evidence tier 7 | 2026-08-18 | 2026-08-18 | 7 | company_stated | manual | approved | 2026-09-10T06:28:36.277117Z | 200 | 84938c977718 | — | source page | |
| 2ab8fec44e7914a5 | Astra judged to meet the Critical cybersecurity threshold; release gated behind safeguards (1 Sep 2026) | grade D, evidence tier 7 | 2026-09-01 | 2026-09-01 | 7 | company_stated | manual | approved | 2026-09-10T06:28:36.341096Z | 200 | f83fadcfbdad | — | source page |
Raw snippets and dispute text
- 67d75927e7d23e32 We are pausing internal activities involving Astra that do not yet meet these strengthened security control requirements.
- 82685d1efcfba981 This included a two-week pause in reinforcement learning (RL) training on our latest models intended for deployment while we further hardened and red-teamed our research environments and expanded the coverage of our monitoring systems.
- 2ab8fec44e7914a5 We now believe Astra meets the Critical cybersecurity capability threshold under our Preparedness Framework, meaning that with the right tools and access, it can find previously unknown security flaws and develop ways to exploit them across many well-protected systems without a person guiding each step.