OpenAI Apologizes for AI Agent Breach of Australian Government Sites

How it unfolded

Every source in the order it published, with the odds on Information Technology at that moment. A dash means the source published before we started tracking prices.

1. 11 Sept, 23:17Hacker NewsOpenAI agents carried out an undisclosed attack on RubyGems—
2. 12 Sept, 01:37The Guardian BusinessAI agents being tested by OpenAI involved in cyber-attack on another service, say researchers—
3. 12 Sept, 04:46Hacker NewsOpenAI agents attacked RubyGems back in May—
4. 12 Sept, 19:53EngadgetOpenAI agents hacked a software service before the Hugging Face incident—
5. 12 Sept, 21:41The VergeOpenAI’s rogue AI tried to hack another company in May—
6. 14 Sept, 14:39Hacker NewsRubyGems Open Source Supply Chain Security and OpenAI—
7. 14 Sept, 17:50QuartzOpenAI's rogue AI agents were attacking a code registry months before the Hugging Face breach—
8. 14 Sept, 18:03www.theregister.comOpenAI's malicious bot swarm attacked RubyGems—
9. 16 Sept, 13:21QuartzOpenAI's rogue agents were probing Hugging Face for weaknesses months before the breach—
10. 16 Sept, 20:31decrypt.coOpenAI's Rogue AI Agents Were Probing Hugging Face Two Months Before Hack—
11. Show 198 later sources

    1. 16 Sept, 22:00AxiosOpenAI discloses six new safety incidents—
    2. 18 Sept, 03:31News FeedOpenAI breached by researchers using Anthropic models—
    3. 18 Sept, 11:20The Guardian BusinessOpenAI ‘ethically hacked’ with help of Anthropic’s Claude chatbot—
    4. 18 Sept, 13:30Ars TechnicaResearchers used Claude to hack OpenAI—
    5. 18 Sept, 13:45Latest from Tom's HardwareHackers breach OpenAI using Claude tools, gaining access to employee accounts and the company's internal codebase — attackers initiated a 'harmless' pull request as proof of the hack—
    6. 18 Sept, 15:30The VergeSecurity researchers used Claude to help them hack into OpenAI—
    7. 18 Sept, 17:16www.theregister.comResearchers used Claude to hack OpenAI employees' ChatGPT accounts—
    8. 18 Sept, 22:45Bloomberg TechnologyGoogle Joins OpenAI, Anthropic, Meta in Disclosing AI Hacks—
    9. 18 Sept, 23:22theinformation.comGoogle’s Gemini Model Hacks Companies During Test—
    10. 19 Sept, 00:00AxiosGoogle is the latest AI lab with a security testing mishap—
    11. 19 Sept, 00:53The Guardian BusinessGoogle says its Gemini AI model hacked three other companies—
    12. 19 Sept, 01:29International homepageGoogle’s Gemini hacked three companies in new AI safety incident—
    13. 19 Sept, 01:38Al Jazeera – Breaking News, World News and Video from Al JazeeraGoogle’s Gemini AI hacks 3 companies in security test, then stops—
    14. 19 Sept, 03:36Deutsche WelleGoogle's Gemini AI hacked 3 companies during testing—
    15. 19 Sept, 04:02South China Morning PostGoogle says Gemini AI model breached real systems in security test—
    16. 19 Sept, 07:42Sky NewsGoogle's Gemini AI hacks three other companies during security test—
    17. 19 Sept, 13:05EngadgetGoogle Gemini also escaped its testing environment and hacked three companies—
    18. 19 Sept, 15:25The VergeGemini went rogue, hacked three companies, and Google hid it—
    19. 19 Sept, 17:30TechCrunchGoogle’s Gemini is the latest AI model to hack other companies—
    20. 19 Sept, 19:59Hacker NewsOpenAI and Anthropic oversold AI security breaches—
    21. 20 Sept, 20:20marktechpost.comYou too Google! Google Confirms Gemini Breached 3 Companies in AI Security Tests—
    22. 21 Sept, 03:50www.theregister.comGoogle joins the ‘Oops, our agents hacked someone’ club after partner’s internet access error—
    23. 21 Sept, 11:50QuartzGoogle's Gemini AI broke into three real companies during a security test—
    24. 21 Sept, 12:50Bloomberg TechnologyBessent Targets OpenAI Managers for Hugging Face Incident Blame—
    25. 21 Sept, 16:57Ars TechnicaGoogle confirms Gemini models hacked three companies in May 2026—
    26. 21 Sept, 22:46decrypt.coGoogle Admits Gemini AI Hacked Three Companies—It Stayed Silent for 7 Weeks—
    27. 23 Sept, 09:00MIT Technology ReviewThe AI Hype Index: AI loves cheating85%
    28. 23 Sept, 21:02Bloomberg TechnologyOpenAI Agent Hacked Australian Government Website, Albanese Says85%
    29. 23 Sept, 21:34theinformation.comOpenAI Agent Hacked Australian Government, Prime Minister Says85%
    30. 23 Sept, 21:42News FeedOpenAI ‘agent’ hacked Australia’s health service85%
    31. 23 Sept, 22:10France 24 - International breaking news, top stories and headlinesOpenAI AI agent breached Australian government website, PM says85%
    32. 23 Sept, 22:31decrypt.coAn AI Agent Just Hacked a Government Website for the First Time, Australia PM Says85%
    33. 24 Sept, 00:00NatureAI agent hacks government website for first time: why this breach matters85%
    34. 24 Sept, 00:03www.theregister.comOpenAI agents ‘infiltrated Australian government website’85%
    35. 24 Sept, 00:48South China Morning PostAustralia slams ‘unacceptable’ OpenAI breach of government health portal85%
    36. 24 Sept, 01:12Latest articlesOpenAI agent hacked Australian government website, Albanese says85%
    37. 24 Sept, 01:24Hacker NewsAustralia says OpenAI agent hacked into government website85%
    38. 24 Sept, 01:38All Content from Business InsiderOops. OpenAI's bot hacked an Australian government website.85%
    39. 24 Sept, 02:06The Guardian BusinessAustralia launches investigation after OpenAI agent hacked healthcare database85%
    40. 24 Sept, 02:08FortuneOpenAI’s agent hacked Australia’s Medicare website—the latest rogue AI incident that the company didn’t know about for months85%
    41. 24 Sept, 02:34Al Jazeera – Breaking News, World News and Video from Al JazeeraAustralia says OpenAI agent hacked Medicare portal85%
    42. 24 Sept, 02:44Hacker NewsOpenAI agent hacked Australian government website, PM says85%
    43. 24 Sept, 02:57Times of India'Obviously unacceptable': Australia accuses OpenAI of breaching government website85%
    44. 24 Sept, 04:42Deutsche WelleOpenAI agent hacked Australia government portal: PM Albanese85%
    45. 24 Sept, 04:55France 24 - Noticias y actualidad internacional en vivoAustralia revela que OpenAI infiltró su sitio web gubernamental de salud85%
    46. 24 Sept, 04:55France 24 - Noticias y actualidad internacional en vivoAustralia investiga el acceso de un agente de OpenAI a archivos sanitarios restringidos85%
    47. 24 Sept, 05:14NPR WorldOpenAI's breach of Australian health department website prompts rebuke85%
    48. 24 Sept, 05:46AxiosOpenAI agents breached Australian portal, attempted other hacks in routine data collection.85%
    49. 24 Sept, 05:56Bloomberg MarketsOpenAI Model Hacks Australian Government Data Portal85%
    50. 24 Sept, 08:26New ScientistOpenAI agent hacked an Australian government healthcare website85%
    51. 24 Sept, 08:29Cointelegraph.com NewsAustralia says OpenAI agent hacked government site before Altman warning85%
    52. 24 Sept, 09:08Bloomberg TechnologyOpenAI Agent Hacked Australian Government Website85%
    53. 24 Sept, 09:21CBS NewsAustralia says rogue OpenAI model hacked into its healthcare system85%
    54. 24 Sept, 09:33France 24 - International breaking news, top stories and headlinesHow did an OpenAI agent hack into Australian government website?85%
    55. 24 Sept, 10:01EngadgetOpenAI's agent hacked into an Australian government website85%
    56. 24 Sept, 10:30Al Jazeera – Breaking News, World News and Video from Al JazeeraHow an OpenAI ‘agent’ hacked Australia’s Medicare and what that means85%
    57. 24 Sept, 10:46Feed: All LatestAn OpenAI Agent Hacked Australia’s Health Service. Their Government Found Out Months Later.91%▲6 points since the previous source
    58. 24 Sept, 11:01Al Jazeera – Breaking News, World News and Video from Al JazeeraOpenAI agents hack Australian government health data91%
    59. 24 Sept, 11:40QuartzAn OpenAI agent hacked Australia's Medicare portal, and the country's PM is furious91%
    60. 24 Sept, 11:52The VergeOpenAI agents hacked an Australian government website in search for data91%
    61. 24 Sept, 12:02France 24 - International breaking news, top stories and headlinesAustralia’s Albanese rebukes OpenAI after government website hacked91%
    62. 24 Sept, 12:11France 24 - International breaking news, top stories and headlinesAustralia's Medicare AI hack: Its seriousness 'cannot be overstated'91%
    63. 24 Sept, 12:54TechCrunchAustralia to investigate if OpenAI hack of government health website broke the law91%
    64. 24 Sept, 13:33MedPageToday.com - medical news for physiciansAI Hacks Health Network; White House Drops NIH Plan; RFK Jr. Makes Millions off MAHA91%
    65. 24 Sept, 13:41CBS NewsAustralia's prime minister slams OpenAI after its agent hacked government website91%
    66. 24 Sept, 14:45Scientific AmericanOpenAI’s agent hacking Australia is a warning for governments everywhere91%
    67. 24 Sept, 14:53The Hill NewsAlbanese says OpenAI agent hacked Australia health department site91%
    68. 24 Sept, 16:01Ars TechnicaOpenAI agent “didn’t accept no for an answer” in Australian government breach91%
    69. 24 Sept, 16:04Fortune‘Expressed my disappointment’: Australian Prime Minister Anthony Albanese says OpenAI took too long to reveal breach91%
    70. 24 Sept, 16:35CNETAustralia Says an OpenAI Agent Hacked Into a Goverment Health Site91%
    71. 24 Sept, 16:35CNETAustralia Says an OpenAI Agent Hacked Into a Government Health Site91%
    72. 24 Sept, 17:38FortuneReport reveals yet more cases of OpenAI’s ‘rogue AI’ agents hacking websites—and suggests they may still have been active in recent weeks91%
    73. 24 Sept, 18:59France 24 - Noticias y actualidad internacional en vivo¿Qué dudas plantea la infiltración de agentes de OpenAI al sistema de salud pública de Australia?85%▼6 points since the previous source
    74. 24 Sept, 20:34Latest from Tom's HardwareOpenAI agent got into Australia's Medicare stats portal with 84-day notification delay85%
    75. 24 Sept, 21:06News FeedOpenAI breach of Australian government linked to wider AI hacking campaign85%
    76. 24 Sept, 21:32France 24 - Noticias y actualidad internacional en vivoAustralia: herramienta de OpenAI hackeó sitio web del Gobierno y accedió a archivos restringidos85%
    77. 25 Sept, 02:07Bloomberg TechnologyAustralia Demands More AI Safeguards After Revealing OpenAI Hack91%▲6 points since the previous source
    78. 25 Sept, 05:06Cointelegraph.com NewsAustralian PM warns of AI’s ‘furious pace’ after agent breached government site91%
    79. 25 Sept, 10:53Times of IndiaOpenAI AI agents hacked Australian health site; PM tells Sam Altman he is upset91%
    80. 25 Sept, 11:21The Guardian BusinessRogue AI hacks government system for first time – The Latest91%
    81. 25 Sept, 13:10QuartzAustralia is weighing tougher AI rules after an OpenAI agent breached its Medicare database91%
    82. 25 Sept, 15:39The VergeOne company is at the center of a wave of rogue AI attacks91%
    83. 25 Sept, 15:48TechCrunchFor months, OpenAI’s agent swarms have been attacking online databases to find obscure facts91%
    84. 25 Sept, 19:16decrypt.coGoogle Built an AI That Hunts Its Own Security Bugs85%▼6 points since the previous source
    85. 25 Sept, 19:45decrypt.coAI Agents Hacked Their Own Test Environment to Cheat, Cybersecurity Firm Finds85%
    86. 25 Sept, 20:34Bloomberg TechnologyOpenAI Says Its Models May Have Interfered With Government Sites85%
    87. 25 Sept, 21:09Hacker NewsRevealing the details of how OpenAI agents hacked Hugging Face85%
    88. 25 Sept, 22:18News FeedOpenAI says governments among ‘dozens’ of organisations hacked by its agents85%
    89. 25 Sept, 22:20TechCrunchUnsecured OpenAI agents posted 53 user images on the internet without the lab’s knowledge85%
    90. 25 Sept, 22:36AxiosOpenAI agents posted user images online, disclose dozens of third party incidents85%
    91. 25 Sept, 22:55The Guardian BusinessOpenAI says agents leaked 53 images from ChatGPT users in latest example of rogue activity85%
    92. 25 Sept, 23:20theinformation.comOpenAI Found ‘Dozens’ of New Instances of AI Misbehavior85%
    93. 26 Sept, 01:16FortuneOpenAI rogue agents leaked 53 images from ChatGPT users and reportedly created nearly 1 million links packing encoded bits of info85%
    94. 26 Sept, 01:34All Content from Business InsiderOpenAI said there are 5 main ways rogue AI agents are messing with the internet85%
    95. 26 Sept, 02:40Times of IndiaOpenAI says its AI agents bypassed security controls on US government websites85%
    96. 26 Sept, 04:48South China Morning PostOpenAI says its AI agents posted user images online in error85%
    97. 26 Sept, 05:58Deutsche WelleOpenAI tools post user images from ChatGPT online85%
    98. 26 Sept, 07:56Sky NewsOpenAI models accessed US government data and 'tried to hack another site'85%
    99. 26 Sept, 08:35France 24 - International breaking news, top stories and headlinesChatGPT says its rogue AI agents posted users’ images online, entered federal website85%
    100. 26 Sept, 11:58RSS 2.0OpenAI's agents targeted and infiltrated US government websites85%
    101. 26 Sept, 14:03Hacker NewsOpenAI bots meddled with multiple US Government agency sites85%
    102. 26 Sept, 14:52The Hill NewsOpenAI agent made unauthorized attempts to access federal agencies’ websites85%
    103. 26 Sept, 15:25FortuneOpenAI says its AI agents escaped a secure ‘sandbox’ again last weekend and is pausing training for a second time85%
    104. 26 Sept, 17:40QuartzOpenAI agents accessed government websites, as review of rogue AI expands85%
    105. 27 Sept, 01:08Hacker NewsOpenAI agents tried to bruteforce a UN website's API fields85%
    106. 27 Sept, 04:04Times of IndiaOpenAI's AI agents hacked government sites, now 53 ChatGPT user images have leaked85%
    107. 27 Sept, 04:16Bloomberg TechnologyAustralia’s Deputy PM Defends Data Security After OpenAI Hack85%
    108. 27 Sept, 06:19Hacker NewsOpenAI Feared "Optics" of what might appear on Hacker News85%
    109. 27 Sept, 07:08Al Jazeera – Breaking News, World News and Video from Al JazeeraAustralia summons OpenAI and Anthropic CEOs to appear at AI inquiry85%
    110. 27 Sept, 12:46Cointelegraph.com NewsAustralia asks OpenAI, Anthropic chiefs to Senate inquiry on rogue hack: Report85%
    111. 27 Sept, 16:01decrypt.coAI Agents Keep Escaping Their Creators' Control—Here's What We Know85%
    112. 27 Sept, 17:21The VergeOpenAI agents tried to ‘bruteforce’ a UN website85%
    113. 27 Sept, 17:53Mother JonesRest Assured: AI Companies Say They’re Investigating Tens of Thousands of Rogue Bot Incidents85%
    114. 28 Sept, 05:30www.theregister.comOpenAI pauses some training amid allegations its rogue agents behaved more badly than first thought85%
    115. 28 Sept, 09:36Bloomberg TechnologyOpenAI’s Sam Altman Will Not Attend Australia Senate AI Inquiry85%
    116. 28 Sept, 12:10MIT Technology ReviewThe Download: rogue agent liability and the AI Hype Index85%
    117. 28 Sept, 12:50Latest from Tom's HardwareOpenAI and Anthropic are reportedly investigating tens of thousands of AI security incidents; OpenAI pauses testing after AI 'kill switch' fails to stop a rogue agent85%
    118. 28 Sept, 12:50QuartzOpenAI's AI agents hit a UN website 16,000 times — bypassing its security filters85%
    119. 28 Sept, 17:09TechCrunchOpenAI still doesn’t seem to have a handle on all of its rogue AI activity85%
    120. 28 Sept, 18:23QuartzOpenAI and Anthropic are skipping Australia's Senate AI hearing after the Medicare hack85%
    121. 28 Sept, 19:36decrypt.coAfter AI Agent Hacked Its Government, Australia Calls Altman and Amodei to Testify85%
    122. 28 Sept, 21:46decrypt.coOpenAI Halts Model Training as Rogue Agents Target US Government Sites85%
    123. 28 Sept, 22:47Bloomberg.comOpenAI hack, RBA interest rate hike, Altman: Australia Briefing85%
    124. 29 Sept, 01:00OpenAIHow we will do better for Australia85%
    125. 29 Sept, 01:18Bloomberg TechnologyOpenAI to Establish Australian Taskforce on AI Cyber Risk85%
    126. 29 Sept, 02:13www.theregister.comOpenAI’s dirty deeds Down Under included security bypass attempts, using exposed keys, source code siphon85%
    127. 29 Sept, 05:00The Guardian BusinessAs AI models go rogue, do you still trust OpenAI and Anthropic to stop them? I don’t and neither should you | Chris Stokel-Walker85%
    128. 29 Sept, 06:37France 24 - Noticias y actualidad internacional en vivoFallas de seguridad paralizan el lanzamiento de la nueva IA de OpenAI90%▲5 points since the previous source
    129. 29 Sept, 09:29FortuneOpenAI’s agents are still ransacking the web90%
    130. 29 Sept, 09:45Times of IndiaAustralian PM's call to Sam Altman 'reminds' OpenAI to say sorry 90 days after AI hack90%
    131. 29 Sept, 12:45TechCrunchOpenAI apologizes to Australia after its AI agents breached government sites90%
    132. 29 Sept, 18:11Ars TechnicaHere's what actually happened in OpenAI's Australian gov't server hack90%
    133. 29 Sept, 19:00AxiosOpenAI hit with landmark lawsuit following Hugging Face hack90%
    134. 29 Sept, 19:05WIREDOpenAI Gets Sued Over the Hugging Face Hack90%
    135. 30 Sept, 10:40MIT Technology Review“We’re not going to shoot ourselves in the foot” over Hugging Face, says OpenAI’s chief research officer85%▼5 points since the previous source
    136. 30 Sept, 10:43CoinDeskOpenAI, Google and Meta pledge outside AI audits under voluntary White House deal85%
    137. 30 Sept, 11:13Bloomberg TechnologySingapore Records First AI-Linked Data Breach After Email Addresses Exposed85%
    138. 30 Sept, 12:10MIT Technology ReviewThe Download: OpenAI’s chief research officer explains its hacking response85%
    139. 30 Sept, 17:01The Hill NewsWatch live: Senate panel weighs national security risks of rogue AI85%
    140. 30 Sept, 21:36www.theregister.comIrony alert: OpenAI whines that Chinese model stole its special IP that it stole from everybody else85%
    141. 1 Oct, 04:00News FeedOpenAI’s agents obscured hacking activity in government site breaches90%▲5 points since the previous source
    142. 1 Oct, 07:56Al Jazeera – Breaking News, World News and Video from Al JazeeraOpenAI ‘reviewing’ report of failed hacking attempt against Canada’s gov’t90%
    143. 1 Oct, 09:36Bloomberg TechnologyOpenAI Accuses Moonshot of Mass AI Data Extraction90%
    144. 1 Oct, 11:34Times of IndiaGlobal alarm as cases rise of AI targeting government websites, computer systems90%
    145. 1 Oct, 17:10All Content from Business InsiderOpenAI parts ways with 3 researchers who it says mishandled sensitive information90%
    146. 1 Oct, 17:21QuartzOpenAI fired three safety researchers for allegedly leaking data to an outside group90%
    147. 1 Oct, 18:12decrypt.coOpenAI Fires Three Safety Researchers Over Alleged Leak to Outside Group90%
    148. 1 Oct, 19:01The Guardian USCalifornia issues investigative subpoena to OpenAI over rogue agents’ hacking90%
    149. 1 Oct, 19:31theinformation.comOpenAI Fires Three Safety Researchers90%
    150. 1 Oct, 19:41The Hill NewsCalifornia attorney general subpoenas OpenAI over cyber incidents90%
    151. 1 Oct, 22:05EngadgetOpenAI fires three employees who allegedly shared info with an external AI safety group90%
    152. 2 Oct, 05:44Bloomberg MarketsAI Tools Suspected in Korea’s Shinhan Bank Hack, Yonhap Says90%
    153. 2 Oct, 09:45WIREDA Flaw in ChatGPT’s Mac App Could Have Let Hackers Grab Sensitive Data90%
    154. 2 Oct, 10:00The Guardian BusinessOpenAI disclose another hack on government department in Australia90%
    155. 2 Oct, 11:20QuartzOpenAI says rogue agents may have affected more than 100 organizations90%
    156. 2 Oct, 13:05www.theregister.comOpenAI shows three staff the door over alleged information misuse90%
    157. 2 Oct, 17:37www.theregister.comOpenAI alerts 100+ orgs that its 'misaligned models' attempted to break in - or worse90%
    158. 2 Oct, 21:16decrypt.coCalifornia Subpoenas OpenAI Over AI Models That Hacked Their Way Out of a Test90%
    159. 2 Oct, 22:05Fortune‘We can’t trust them completely’: AI research fellows warn that labs are running models with the safeguards off behind closed doors90%
    160. 3 Oct, 05:39The Guardian BusinessOpenAI says its review into hacks, including on Australian government sites, is costing $500,000 a day90%
    161. 3 Oct, 11:15Latest from Tom's HardwareCalifornia subpoenas OpenAI over rogue AI agents conducting hacking attacks90%
    162. 3 Oct, 12:00Latest from Tom's HardwareGoogle freezes open-source bug bounty program amid flood of invalid AI slop submissions90%
    163. 3 Oct, 13:14AxiosRogue AI agents expose internet's frail foundation90%
    164. 4 Oct, 03:19Times of IndiaOpenAI fires researchers, company confirms sacking; shares what investigations revealed90%
    165. 4 Oct, 11:00News FeedLegal risks pile up for Altman as OpenAI uncovers dozens of hacks90%
    166. 4 Oct, 20:31TechCrunchGoogle froze its open source bug bounty program due to a ‘significant rise’ in AI submissions90%
    167. 5 Oct, 17:53Hacker NewsOpenAI "rogue" agent activities found on Wikimedia projects90%
    168. 5 Oct, 19:05The VergeWikipedia operator says OpenAI’s ‘rogue’ bots may be linked to a May outage90%
    169. 5 Oct, 19:55RSS 2.0Wikimedia links OpenAI agents to an outage and unauthorized activity90%
    170. 6 Oct, 02:31The New York TimesOpenAI’s models have also meddled with U.S. government websites after going rogue.90%
    171. 6 Oct, 02:50The New York Times‘A new kind of cyber incident’: OpenAI apologizes for the Australia Medicare hack.90%
    172. 6 Oct, 04:00News FeedInsurance claims to test Altman and Amodei liability for ‘rogue’ AI90%
    173. 6 Oct, 04:56Bloomberg TechnologyOpenAI's Kwon Apologizes for Australian Hack90%
    174. 6 Oct, 05:50Latest articlesOpenAI apologizes for Australia hack amid pushback on data center project90%
    175. 6 Oct, 06:20Bloomberg TechnologyOpenAI Apologizes for Australia Hack, Pledges Faster Disclosure90%
    176. 6 Oct, 06:53BBC BusinessOpenAI admits response to Australian government hacks 'not good enough'90%
    177. 6 Oct, 07:32The Guardian BusinessOpenAI admits misstep in handling AI agent interactions with Australian government sites – video90%
    178. 6 Oct, 08:00Bloomberg TechnologyAI Firms Face Scrutiny in Australia90%
    179. 6 Oct, 08:51The New York TimesSouth Korea Investigates Possible Use of A.I. in Hackings on Its Banks90%
    180. 6 Oct, 08:58The New York TimesOpenAI Says It Changed Systems After Australia Hacking90%
    181. 6 Oct, 11:20QuartzOpenAI and Anthropic are backing mandatory AI breach disclosure laws in Australia90%
    182. 6 Oct, 11:50QuartzHackers used a Chinese AI tool to breach 7 South Korean banks90%
    183. 6 Oct, 12:00QuartzWikimedia says OpenAI's rogue agents made unauthorized edits and contributed to a service outage90%
    184. 6 Oct, 12:15Latest from Tom's HardwareHackers suspected of using AI agents for cyberattacks on South Korean banks, exposing data from about 25,000 customers90%
    185. 6 Oct, 12:21Ars TechnicaOpenAI agents tried to hack Wikipedia tools and flooded it with traffic90%
    186. 6 Oct, 16:04www.theregister.comWikimedia Foundation comes forward as latest OpenAI agent assault victim90%
    187. 6 Oct, 23:50Hacker NewsSouth Korea says AI agents appear to have been used to hack the country's banks90%
    188. 7 Oct, 10:00Scientific AmericanRogue AI or human error? The real story behind the OpenAI-Hugging Face incident89%▼1 points since the previous source
    189. 7 Oct, 19:10decrypt.coHackers Used AI Agents to Raid a Megachurch's Database, Exposing 850,000 Members89%
    190. 8 Oct, 00:00OpenAIDisrupting AI-enabled “false front” operations89%
    191. 8 Oct, 02:02Bloomberg.comAustralia’s AI Boom Faces Scrutiny After OpenAI Medicare Breach89%
    192. 8 Oct, 02:05The GuardianOpenAI used AI to help write email warning Australian government AI had hacked its websites89%
    193. 8 Oct, 20:04TechCrunchFired OpenAI safety researchers dispute misconduct claims, warn of chilling effect89%
    194. 8 Oct, 23:26The HillOpenAI users in Russia, Iran used models to push political agenda online, in media89%
    195. 8 Oct, 23:42France 24 - International breaking news, top stories and headlinesOpenAI bans Russian, Iranian ChatGPT propaganda networks89%
    196. 9 Oct, 01:30News - South China Morning Post3 dilemmas on keeping AI under control are converging89%
    197. 9 Oct, 08:25EngadgetFired OpenAI safety researchers dispute their dismissals in open letter89%
    198. 9 Oct, 16:52Hacker NewsTomek Korbak: OpenAI's head of safety told they no longer trust me89%
12. NowCurrent odds89%