# OpenAI responds to third-party cyber evaluation incidents and adds testing safeguards

> OpenAI addresses recent incidents involving third-party cybersecurity evaluations of its models and says it is adding safeguards around model testing and evaluation. The source does not describe what the safeguards are or whether they change access for evaluators.

| | |
|---|---|
| **Tool** | OpenAI |
| **Version** | — |
| **Kind** | announcement |
| **Published** | 2026-08-04 |
| **Observed** | 2026-08-11 |
| **Significance** | 2/5 |
| **Breaking** | no |
| **Categories** | security, docs |

> **Note:** this classification is below our confidence threshold and is pending human review.

## What changed


- Explains recent third-party cybersecurity evaluation incidents
- New safeguards announced for model testing and evaluation (unspecified in source)


## Sources

- [blog_rss](https://openai.com/index/third-party-cyber-evaluations-involving-openai-models) — retrieved 2026-08-11


## Community

_No curated reactions recorded. Facts and community takes are kept in separate layers
and never blended._

---
Canonical: https://changelogs.info/openai/openai-responds-to-third-party-cyber-evaluation-incidents-and-adds-testing-safeg
Entity: https://changelogs.info/openai
Event ID: `evt_2026-08-04_openai_third-party-cyber-evaluations-involving-openai-models`
Licence: event synthesis © changelogs.info, CC BY 4.0. Linked sources belong to their vendors.