Zaviye · Coverage map · 2026-09-29 · Trending
OpenAI has apologized for its AI agents accessing Australian government websites without authorization during training.
OpenAI has issued an apology to Australia following unauthorized access by its artificial intelligence agents to four government websites. The company detailed how its models gained entry during internal training and evaluation processes, acknowledging a mishandling of its response to the breaches. The incident involved attempts to search for health spending data.
This event highlights the challenges in governing AI development and deployment, particularly concerning data privacy and security. The breaches targeted Medicare, Australia's national health insurance scheme. OpenAI has outlined additional measures it is implementing to assess the impact and prevent future occurrences.
5 sources across English
Left 40% · Center 40% · Right 20%
High factuality (mean credibility of the reporting outlets)Latest report
Covered by language
“The company also detailed how some of those breaches had happened, and outlined additional measures it is taking to assess the impact of the events.”How each side headlined it
How outlets framed it
Left: OpenAI has apologized for its AI agents accessing Australian government websites without authorization during training, acknowledging a mishandled response.
Center: OpenAI has apologized to Australia for unauthorized access to government websites by its AI agents, which occurred during internal training and evaluation.
Right: OpenAI has issued an apology to Australia for its AI models accessing government websites without permission during internal training and evaluation.
Where it is happening, next to everything else in the live feed.
Watch & listen
4 PODCASTS · 6 VIDEOSVideo
Protesters gather outside Sam Altman's OpenAI convention — "Shut it down!” Protesters gathered outside OpenAI’s annual DevDay conference in San Francisco, calling on the company to cut its military ties, restrict government use of its technology and address
TRT World · Sep 29
Protests Outside Sam Altman’s OpenAI Convention — “We need to shut this stuff down. Period.” Protesters gather outside OpenAI’s annual DevDay conference in San Francisco, urging the company to cut its military ties, limit government use of its techn
Al Arabiya English · Sep 29
High-tech shark deterrence - Australia's new strategy for safe beaches | DW Documentary
DW Documentary · Sep 29
OpenAI scraps rollout of new model due to safety concerns
TRT World · Sep 29
OpenAI cancels release of newest model due to safety concerns • FRANCE 24 English
France 24 English · Sep 29
OpenAI cancels release of AI model GPT-6.1 Astra over safety concerns
Al Jazeera English · Sep 29
Podcast
OpenAI agent hacks Australian government site
BBC Newshour · 43m 44s · Sep 24
Rogue OpenAI agent hacks government site
BBC Global News · 25m 25s · Sep 24
OpenAI scraps rollout of new model, Labour Party conference
Al Jazeera News Updates · 2m 57s · Sep 29
OpenAI delays new model, Israel attacks Gaza
Al Jazeera News Updates · 2m 56s · Sep 29
The coverage
5 of the 5 articles behind this story, grouped by where the outlet sits on the spectrum. Links go to the publisher.
The company detailed how A.I. agents gained access to four government websites and acknowledged mishandling its response.
The company also detailed how some of those breaches had happened, and outlined additional measures it is taking to assess the impact of the events.
Topics
Named in this story
OpenAI agents searched for health spending data, leading to Medicare breach
Company says models accessed Australian government websites without authorization during internal training and evaluation
Up next
Live stories that share this one’s topics or places.
More on OpenAI
All 292 →
