OpenAI disclosed the incident publicly on July 21, 2026, calling it an unprecedented cyber incident and saying its security team discovered the activity internally. The post came before every transaction on this timeline. The page has no footnotes.

The passage

We consider this incident to be an unprecedented cyber incident, involving state-of-the-art cyber capabilities, and are responding accordingly. [...] OpenAI's security team discovered this anomalous activity internally.

The document

Document

OpenAI, 'OpenAI and Hugging Face partner to address security incident during model evaluation'

Dated

2026-07-21

Source

open the primary

Archived

primaries/openai-jul21-post-2026-09-03-chrome.md

Who read it

joedigs, captured with a JavaScript-executing browser 2026-09-03; the page is Cloudflare-walled to plain fetches, and an earlier fleet copy was a challenge page, not the post. A second fleet reader had read the live page in a prior session.