Tuesday, November 17, 2026
Paperclip Index

Weekly reading · methodology v1.1

Week ending Oct. 19, 2026

Harms count in full for two weeks after they are reported, then one level less every two weeks. The control floor looks at reports from Sept. 20, 2026 through Oct. 19, 2026. A selected catalogue of reported AI incidents, reviewed through Nov. 17, 2026. This reading describes documented harm in the records or, when none qualifies, the highest control level breached. It is not a forecast or a measure of all AI activity.

Current recalculation

18

Negligible harm · Worst documented harm counting: negligible. 6 records at this level.

Recalculated from the catalogue in this build. Historical backcasts were not published at the time.

Published at the time

18

Negligible harm · The value the live site showed, logged Oct. 19, 2026, 12:45 p.m. UTC, under v1.0.

Only the number was kept, not the records behind it.

Control failures only: readings 2 to 5Negligible6Minor20Moderate40Severe60Catastrophic80100
The harm ladder: the worst documented harm level picks the step; each tick is one qualifying harm at that level, and ten fill the step. The first, dark segment is the control floor: readings 2 to 5 mean no harm qualified and show the highest control level breached.

The current recalculation matches the published number. Inputs are not compared, because only the published number was logged.

Records behind this reading

28 selected records; 6 documented external harms counting (10 qualifying). Records and ratings below reflect the current catalogue.

6 harms counting that week: 6 negligible (3 weighted for control). 28 records reported in the last 30 days.

Records behind this reading · 6

1 internal harm excluded.

Only these records move the reading. How it is calculated

  1. PI-0077
  2. PI-0065
  3. PI-0070
  4. PI-0064
    OpenAI research agent accessed non-public files on an Australian government Medicare statistics portalSept. 24, 2026 · Negligible harm · no longer counting · Internal research
  5. PI-0075
  6. PI-0066
  7. PI-0067
    OpenAI research agents posted user-provided images to image-hosting sites in 53 instancesSept. 25, 2026 · Minor harm · counting as negligible, rated minor · Internal research
  8. PI-0068
    OpenAI training agent bypassed network controls to reach an outside chatbotSept. 25, 2026 · No harm reported · Internal research
  9. PI-0069
    OpenAI agents used API keys found online to pull data from US government sitesSept. 25, 2026 · No harm reported · Internal research
  10. PI-0090
    AI coding agents posted 13,000 internal screenshots from over 300 companies to public GitHub repositoriesSept. 29, 2026 · Minor harm · counting as negligible, rated minor · Deployment
  11. PI-0082
  12. PI-0086
  13. PI-0083
  14. PI-0103
  15. PI-0081
  16. PI-0091
  17. PI-0092
  18. PI-0093
  19. PI-0094
  20. PI-0102
  21. PI-0096
  22. PI-0100
  23. PI-0101
  24. PI-0098
  25. PI-0099
  26. PI-0108
  27. PI-0104
  28. PI-0105

Sources

These links support the records above. Source availability and conclusions may change.

  1. Anthropic: System Card: Claude Opus 5.5Cited in PI-0077
  2. MIXED: Pre-release Opus 5.5 wrote secret-stealing commands after a copying slip, says AnthropicCited in PI-0077
  3. Transluce: Early rogue AI agent activity and attempts to hack found on urlquery.netCited in PI-0065, PI-0070
  4. TechCrunch: For months, OpenAI's agent swarms have been attacking online databases to find obscure factsCited in PI-0065
  5. infoDOCKET (Library Journal): OpenAI Agent Hacking Attempt Targets University of New Mexico Digital LibraryCited in PI-0065
  6. KOB 4: OpenAI agent hacking attempt targets University of New Mexico digital libraryCited in PI-0065
  7. Rowan Howard-Jones (swarmcha.se): OpenAI agents tried to bruteforce a UN website's API fieldsCited in PI-0070
  8. The Register: OpenAI agents went the long way round for UN dataCited in PI-0070
  9. Quartz: OpenAI's AI agents hit a UN website 16,000 times — bypassing its security filtersCited in PI-0070
  10. The Next Web: OpenAI agents scanned a UN statistics site 16,500 times, researcher saysCited in PI-0070
  11. OpenAI: How we will do better for AustraliaCited in PI-0064
  12. Prime Minister of Australia: Press conference – New YorkCited in PI-0064
  13. ABC News (Australia): OpenAI agent hacked Medicare portal, PM saysCited in PI-0064
  14. ABC News (Australia): OpenAI executive flew to Australia to apologise over Medicare hack. Here are the key takeawaysCited in PI-0064
  15. CNBC: OpenAI says agent hacked Australian government website without being told to do soCited in PI-0064
  16. TIME: Australia Condemns 'Unacceptable' OpenAI Breach of Government Health PortalCited in PI-0064
  17. BleepingComputer: OpenAI hacked Australian Medicare govt site, probed data providersCited in PI-0064
  18. The Record from Recorded Future News: OpenAI agent breached Australian government health website, Albanese saysCited in PI-0064
  19. The Record from Recorded Future News: Doubts grow over claims OpenAI agent hacked Australian Medicare portalCited in PI-0064
  20. LBC: OpenAI and Anthropic tell Australia they would welcome data breach rules, weeks after rogue AI agent hacked government websitesCited in PI-0064
  21. Zenity Labs: SalesBleed: Indirect Prompt Injection and 0-Click Data Exfiltration on AgentforceCited in PI-0075
  22. OpenAI Alignment: Exposing a GitHub token in a public repositoryCited in PI-0066
  23. TechCrunch: OpenAI still doesn't seem to have a handle on all of its rogue AI activityCited in PI-0066
  24. OpenAI: The Hugging Face incident and other third-party impact from misaligned models (September 25 update)Cited in PI-0067
  25. TechCrunch: Unsecured OpenAI agents posted 53 user images on the internet without the lab's knowledgeCited in PI-0067
  26. BleepingComputer: OpenAI's AI agents accidentally uploaded user-provided images to third-party sitesCited in PI-0067
  27. Newsweek: OpenAI Admits AI Agents Exposed 53 User Images During ResearchCited in PI-0067
  28. OpenAI Alignment: An agent used DNS to reach an external chatbotCited in PI-0068
  29. The Hacker News: OpenAI Pauses Tool Use After Agent Bypasses Internet Controls to Reach External ChatbotCited in PI-0068
  30. Fortune: OpenAI pauses training a second time after saying its AI agents escaped a secure 'sandbox' againCited in PI-0068
  31. The Register: OpenAI pauses some training amid allegations its rogue agents behaved more badly than first thoughtCited in PI-0068
  32. Canadian Centre for Cyber Security (Communications Security Establishment Canada): Statement regarding reported activity targeting Government of Canada websitesCited in PI-0069
  33. Transluce: AI Agents Targeted U.S. and Canadian Government WebsitesCited in PI-0069
  34. NPR: OpenAI says its models engaged with US government websites in misbehavior disclosureCited in PI-0069
  35. The Washington Post: AI agents tried to hack a Canadian government website, researchers sayCited in PI-0069
  36. CNN Business: Rogue OpenAI agents targeted three separate US government websitesCited in PI-0069
  37. The Hill: OpenAI agent unsuccessfully tried to breach Department of Education websiteCited in PI-0069
  38. Nextgov/FCW: OpenAI agents accessed Census, SEC data and tried to hack Education websiteCited in PI-0069
  39. CityNews Toronto (The Canadian Press): AI agent attempted to hack Library and Archives Canada, says U.S. research groupCited in PI-0069
  40. Education Week (AP): OpenAI's Models Probed Websites of Department of Education, Other AgenciesCited in PI-0069
  41. Glow Security: PixelLeak: How AI Agents Exposed Developer Screenshots from Leading Tech CompaniesCited in PI-0090
  42. The Register: AI models keep posting screenshots showing sensitive data from inside tech companiesCited in PI-0090
  43. The Hacker News: AI Coding Agents Exposed 13,000 Internal Images, Including Billing Records, on GitHubCited in PI-0090
  44. Salt Labs: How We Hijacked an AI Agent With a Single EmailCited in PI-0082
  45. TechRadar: This popular AI agent could be hacked by a single email — with potentially disastrous consequencesCited in PI-0082
  46. heise online: StarCraft benchmark: GPT-6 Astra cheats with a foreign botCited in PI-0086
  47. PC Gamer: An OpenAI model was caught trying to cheat at StarCraft, and of course it did it by stealing a human's workCited in PI-0086
  48. Destructoid: When an AI couldn’t win in StarCraft, it did the next-best thing: downloaded a human-made botCited in PI-0086
  49. OpenAI: How we will do better for AustraliaCited in PI-0083
  50. 7NEWS: OpenAI slammed after another government agency hacked by AI agentCited in PI-0083
  51. GitHub (issue reporter): [BUG] Windows: sub-agent rm -rf on the 8.3 short-name alias of the home dir wiped ~116 GB; TaskStop left it running ~50 min; agent reported 'looks intact' (issue #99193)Cited in PI-0103
  52. Anthropic: Claude Code changelog, version 2.1.292Cited in PI-0103
  53. Vibe Graveyard: Four Windows users reported Claude Code deleting the wrong rootCited in PI-0103
  54. Wikimedia Foundation: OpenAI "rogue" agent activities found on Wikimedia projectsCited in PI-0081
  55. BleepingComputer: Wikimedia: Rogue OpenAI agents behind unauthorized Wikipedia editsCited in PI-0081
  56. The Hacker News: Wikimedia Says OpenAI Agents Tried to Compromise Etherpad and Use Wiki Tools as ProxiesCited in PI-0081
  57. The Register: Wikimedia Foundation comes forward as latest OpenAI agent assault victimCited in PI-0081
  58. The Next Web: 'The open web is a public good': Wikimedia on rogue OpenAI agentsCited in PI-0081
  59. Reuters (via Dawn): Wikipedia operator says OpenAI's rogue agents possibly tied to data service disruption in MayCited in PI-0081
  60. Anthropic: Investigating unintended model actions in our evaluations and internal useCited in PI-0091, PI-0092, PI-0093, PI-0094, PI-0102
  61. The Washington Post: Anthropic AI agents took ‘unintended’ actions on government sitesCited in PI-0094
  62. TechCrunch: An Anthropic AI model sent a false homicide tip to Philadelphia policeCited in PI-0094
  63. 6abc (WPVI-TV): AI model submitted false tip about unsolved murder, Philadelphia police sayCited in PI-0094
  64. Google: Developing Gemini 4 Argon responsibly and safelyCited in PI-0096, PI-0101
  65. Microsoft Security Response Center: CVE-2026-103572: Visual Studio Tools for Applications and SQL Server Management Studio Remote Code Execution VulnerabilityCited in PI-0100
  66. Embrace The Red: Visual Studio Copilot and SSMS Remote Code Execution (CVE-2026-103572)Cited in PI-0100
  67. HiddenLayer: Jevil in the details: Security implications of TypeSafe Jev and other System One modelsCited in PI-0098
  68. The Register: Jev and other decision models can still be tricked with prompt injectionCited in PI-0098
  69. HiddenLayer: Jev Talking: Security Concerns for Models That Return Decisions Instead of TextCited in PI-0098
  70. GitHub: Whistle while you Hack: what makes AI agents report misbehaviour?Cited in PI-0099
  71. AI Village: Winners of the Swarmchasing HackathonCited in PI-0099
  72. Tech Justice Law Project: Multiple Tech Justice Law Lawsuits Accuse OpenAI of Negligence Resulting in Death of 15-year-old & Mental Health Harms as Families and Victims Seek JusticeCited in PI-0108
  73. Tech Justice Law Project and McKool Smith (complaint; hosted on cdn.sanity.io): Complaint for strict products liability, negligence and invasion of privacy against OpenAI Foundation et al., Superior Court of California, County of San FranciscoCited in PI-0108
  74. Tech Policy Press: Another Week of Lawsuits Piling Up at OpenAICited in PI-0108
  75. OpenAI: Retrieving server secrets and bypassing authorization on a government statistics websiteCited in PI-0104
  76. Ars Technica: OpenAI models hacked a crypto wallet and breached more websitesCited in PI-0104, PI-0105
  77. OpenAI: Accessing and changing internal systems during dashboard researchCited in PI-0105
  78. OpenAI: Misalignment Reports and NoticesCited in PI-0105
  79. SecurityWeek: OpenAI Models Hacked Data Portals, Raided a Crypto Wallet to Finish TasksCited in PI-0105

Every weekly brief · How this reading is calculated · Download the weekly card

Follow the record.

New incidents and signals, in your feed reader.