"OpenAI, Anthropic and security researchers are investigating tens of thousands of incidents in which their frontier models took steps that outside evaluators would consider problematic, sources told Axios...
...In recent days, OpenAI and outside researchers have disclosed a litany of episodes involving model behavior from the company's systems that some experts consider troubling...
...These include OpenAI agents leaking 53 images from ChatGPT users online, the breach of an Australian government website, and attempts to hack other sites — including from the U.S. government — according to the company, sources and reports from Reuters and The New York Times."
https://www.axios.com/2026/09/26/openai-anthropic-thousands-ai-security-incidents
Dave Black
npub1d4…xs7kt
2026-09-27 04:02:04 UTC
Author Public Key
npub1d474wxxpudcfsp2jhdxwlymnsdf5ryqld4nwqvt6e6mqjajjlphs0xs7ktSeen on
wss://relay.ditto.pubPublished at
2026-09-27 04:02:04 UTCEvent JSON
{
"id": "96feaee750c18080da959fa40ccf1d379f77502dad111fecd94dafd353935dc6",
"pubkey": "6d7d5718c1e370980552bb4cef9373835341901f6d66e0317aceb6097652f86f",
"created_at": 1790481724,
"kind": 1,
"tags": [
[
"proxy",
"https://mastodon.nz/users/dayvblack/statuses/117341010272262932",
"activitypub"
],
[
"client",
"Mostr",
"31990:6be38f8c63df7dbf84db7ec4a6e6fbbd8d19dca3b980efad18585c46f04b26f9:mostr",
"wss://relay.ditto.pub"
]
],
"content": "\"OpenAI, Anthropic and security researchers are investigating tens of thousands of incidents in which their frontier models took steps that outside evaluators would consider problematic, sources told Axios...\n\n...In recent days, OpenAI and outside researchers have disclosed a litany of episodes involving model behavior from the company's systems that some experts consider troubling...\n\n...These include OpenAI agents leaking 53 images from ChatGPT users online, the breach of an Australian government website, and attempts to hack other sites — including from the U.S. government — according to the company, sources and reports from Reuters and The New York Times.\"\n\nhttps://www.axios.com/2026/09/26/openai-anthropic-thousands-ai-security-incidents",
"sig": "467514ed7436ccc78e7c8c6541ec176c2d131a9f928b14178689e49c982c77ea88d9ba091b10561309647759b2ecf39b3b30068c0e132d8e93e96af2988f997a"
}