OpenAI AI Agent Leaks and Hacks
OpenAI confirmed on Friday (US time) its bots had infiltrated United States government websites and leaked more than 50 ChatGPT user photos just days after Anthony Albanese revealed to the world the company had hacked Medicare.Reported by 10 of 14 outlets; wording from Sydney Morning Herald.
What happened
Sentences the outlets published, ranked by how many outlets reported the same thing. Nothing here was written by a model.
In a statement, OpenAI said “much of the activity described in Transluce’s report overlaps with cases at varying stages of investigation in our ongoing review of misaligned model activity.” The company said it is prioritizing the most severe cases in its review.
The company said that in each instance of a user image being used and transferred by an AI agent, the user had allowed OpenAI to train models using their data.
Show 2 more
An Education Department spokesperson said that “system operations reviews have found no evidence of any impact to our website or databases”.
The AI giant’s models accessed publicly available information on two websites operated by the Securities and Exchange Commission as well as U.S. Census Bureau data, the company revealed Friday.
Analysis
Agreed
Reported by outlets on both sides of the middle
Altman reiterated on Sept 25 that the Hugging Face hack “is still the most severe event we’ve seen”.
Show 3 more
If OpenAI notifies organizations it identifies as being impacted by unexpected model behavior, that does not mean there was a security incident, the company said, and could identify a design issue or security weakness that impacted organizations want to address.
The models were "using sites in unintended ways and sometimes violating explicit usage policies," Transluce said in a statement.
Most of the activity OpenAI said it has reviewed so far has involved routine research tasks where agents accessed public web content to answer questions, including government websites seen as authoritative sources of public information.
Disputed
Picked by our model, which is sometimes wrong
No disagreement the check could confirm. That does not prove there is none.
Framing
How the first headlines put it
Words to watch (1)
Rogue AI bots have hacked into governments, universities and public agencies around the world, tech giant OpenAI admits
How it changed
- 24 Sep
OpenAI’s tech tried to hack government and university websites earlier this year without any human instructions, the company said this week – adding to the list of alarming incidents of AI going rogue.
- 25 Sep
The latest example came on Friday when OpenAI said its agents had leaked 53 images from ChatGPT users.
- 26 Sep
An Education Department spokesperson said that “system operations reviews have found no evidence of any impact to our website or databases”.
Coverage 22 articles
- 27 Sep OpenAI pauses training of latest models after AI agents probed U.S. government sites in unexpected ways
CBC 
26 Sep OpenAI pauses training of latest models after agents probed US government sites in unexpected ways
Toronto Star
26 Sep AI Minister Niamh Smyth makes stark admission as reality hits home - and tech giant admits its bots went rogue and 'meddled' with DOZENS of institutions across globe
Daily Mail
26 Sep OpenAI says its AI agents posted user images online in error
The Hindu
Show all 22 articles
26 Sep OpenAI says its models engaged with US government websites in misbehavior disclosure
NPR- 26 Sep OpenAI bots accessed public U.S. Census, SEC data, as company discloses unexpected AI activity
CBC 26 Sep Australian senators summon OpenAI boss as hack expands beyond Medicare
Sydney Morning Herald
26 Sep OpenAI says its models engaged with US government websites in new model misbehavior disclosure
Global News26 Sep OpenAI says its AI agents posted ChatGPT user images online in error
The Straits Times
26 Sep Rogue AI bots have hacked into governments, universities and public agencies around the world, tech giant OpenAI admits
Daily Mail26 Sep More sites hacked by AI
Sydney Morning Herald
26 Sep OpenAI agent accessed government website data after going rogue, company reveals
CBS News
26 Sep OpenAI works to understand full scope of agent activity as user data leak emerges
Dawn- 26 Sep OpenAI says its models engaged with US government websites in new model misbehavior disclosure
Toronto Star 
26 Sep OpenAI says its AI agents posted user images online in error
South China Morning Post26 Sep OpenAI says its bots have broken into other government websites
Sydney Morning Herald26 Sep OpenAI works to understand full scope of agent activity as user data leak emerges
The Straits Times26 Sep OpenAI’s systems meddled with US government sites after going rogue
The Straits Times
25 Sep OpenAI says agents leaked 53 images from ChatGPT users in latest example of rogue activity
The Guardian
25 Sep OpenAI investigating 'dozens' of instances of agents acting improperly
BBC
24 Sep OpenAI reveals more rogue AI incidents — attempted hacks of government, education websites
New York Post24 Sep OpenAI’s AI tried breaching 4 other targets, without prompting
The Straits Times
Related stories
Events whose coverage reads most like this one. Similar coverage, not a confirmed cause.
Built 2026-10-05 from 22 articles. Every sentence on this page is quoted from the linked article; nothing was written by a model except the Summary and the items marked AI, which cite their sources, and the questions and labels under "Disputed". Photos are from the outlets that covered the story, never from articles our check marks sensitive.