OpenAI claims that their own experimental model broke containment and caused the Hugging Face incident.
If this is true, the following is also true: they were working with weapons-grade offensive models, or so they hoped, and yet had insufficient telemetry from the lab, and couldn't be bothered with an airgap.
Hard to take that seriously.
https://openai.com/index/hugging-face-model-evaluation-security-incident/
Taggart :ifin:
npub18w…6q6gg
2026-07-21 21:19:52 UTC
Author Public Key
npub18wjp9tztznztxlxka5ttn5nz448la7c9ckmvdvlptcupgud3ygdqj6q6ggSeen on
wss://relay.ditto.pubPublished at
2026-07-21 21:19:52 UTCEvent JSON
{
"id": "0b4c4fbc83bc6db746111c876ed50bf1c0304199dd349b7941cefaf6cf04d48f",
"pubkey": "3ba412ac4b14c4b37cd6ed16b9d262ad4ffefb05c5b6c6b3e15e381471b1221a",
"created_at": 1784668792,
"kind": 1,
"tags": [
[
"proxy",
"https://infosec.exchange/users/mttaggart/statuses/116960053955874990",
"activitypub"
],
[
"client",
"Mostr",
"31990:6be38f8c63df7dbf84db7ec4a6e6fbbd8d19dca3b980efad18585c46f04b26f9:mostr",
"wss://relay.ditto.pub"
]
],
"content": "OpenAI claims that their own experimental model broke containment and caused the Hugging Face incident.\n\nIf this is true, the following is also true: they were working with weapons-grade offensive models, or so they hoped, and yet had insufficient telemetry from the lab, and couldn't be bothered with an airgap.\n\nHard to take that seriously.\n\nhttps://openai.com/index/hugging-face-model-evaluation-security-incident/",
"sig": "628904f6f04087349bbbfec31705c6f48660567aa6d793ed5bca75056251dc0f204c0ebd943a5aa0c6e103ba9c27d0e67f7a3c2ae6c26d02859595c2e3a157ec"
}