# Simon Willison — Third-party cyber evaluations involving OpenAI models

- Company: Simon Willison (simonwillison.net)
- Announced: 2026-08-05T23:45:32+00:00
- Category: safety-policy-update
- Subject: Commentary
- Source: https://simonwillison.net/2026/Aug/5/third-party-cyber-evaluations/#atom-everything
- Record: https://forck.live/items/3578-third-party-cyber-evaluations-involving-openai-models

Simon Willison's blog post links to OpenAI's coverage of two third-party cyber evaluation incidents involving OpenAI models, one by the UK AI Safety Institute and another by Irregular, and notes Anthropic's write-up about Irregular's misconfigured environment affecting Claude.

## Evidence

Verbatim from https://simonwillison.net/2026/Aug/5/third-party-cyber-evaluations/#atom-everything:

> This post from OpenAI covers both the UK AI Safety Institute attack (see my previous post) and another attack enabled by Irregular

---

Record: https://forck.live/items/3578-third-party-cyber-evaluations-involving-openai-models
Catalogue: https://forck.live/llms.txt
Feed: https://forck.live/feed.md