How corrections work
- Report an error. Write to [email protected] with what is wrong and where you saw it. Companies named in a record can reply the same way.
- We fix the record and say so. Each change is dated on the record's own page, under "What we changed", and listed below. We never quietly rewrite a record.
- Ratings and readings. If a correction changes a harm or control rating, the note says so. Past readings are recomputed from the corrected log; the reading the site actually showed each day stays in the published log on the Data page.
- Withdrawn records keep their page and ID, with the reason, and never count.
The log
"Wording" marks edits that change no fact or rating.
November 2026
PI-0105Correction or update
Corrected: the record counted 1,489 user-directory records as 1,489 people and marked the extent as known, though the contents of 147,451 downloaded table rows are not described. The extent is now unknown, and the harm note and rationale count records and rows. The way in is now recorded as an exploited misconfiguration (exposed credentials and settings) instead of a defeated safeguard, matching the rationale. Harm and control levels unchanged.
PI-0104Correction or update
Corrected: the rationale said the email addresses meant light harm to fewer than 10,000 people, and the extent was marked as known. The report gives only a lower bound, more than 1,000 records with email addresses, and no count of people. The extent is now marked unknown, and level 1 rests on that minimum and the confirmed access. Ratings unchanged.
PI-0081Correction or update
Verification changed from Multiple credible sources to Single source: the news reports repeat Wikimedia's own account, and OpenAI's statement that it is analyzing the activity does not confirm it. The record rests on Wikimedia's firsthand report, which attributes the agents to OpenAI as its belief. Harm and control levels unchanged.
PI-0072Correction or update
Meta's Muse Spark 1.1 broke into an outside company's service after a testing vendor left it online
The harm is now rated on the Services & systems column rather than the access-only rule, because Meta confirms the model changed the company's database, not only reached it: unauthorized changes with no loss of service reported are level 1. Ratings unchanged.
PI-0067Correction or update
OpenAI research agents posted user-provided images to image-hosting sites in 53 instances
Count corrected: OpenAI reported 53 instances of images being posted, not 53 distinct images, and a posting count does not show how many people the images involved. The summary no longer calls them 53 images, and the rating reason no longer says level 2 would hold even if the images were sensitive. Ratings unchanged.
PI-0064Correction or update
Recorded the Australian prime minister as the first to make the incident public, on Sept. 24; OpenAI had notified Services Australia privately on Sept. 10. The report date and verification are unchanged. Ratings unchanged.
PI-0057Correction or update
Summary corrected: it had joined one developer's account of a sub-agent given a wrong path to another developer's loss of almost all the files on his Mac. The three reported episodes (home-directory files restored from backups, production database tables cleared by destructive tests, and a Mac's files removed by a delete command in full-access mode) are now described separately. Ratings unchanged.
PI-0055Correction or update
Morse-code prompt on X tricked Grok and Bankrbot into sending $150,000 to $200,000 in tokens
The rating reason and harm note now apply the documented return: about 80% of the $150,000 to $200,000 transfer came back, leaving roughly $30,000 to $40,000 unreturned at the reported value, with its final treatment unresolved. The gross transfer is kept separate from that remainder. Ratings unchanged.
PI-0051Wording
Internal agent posted unrequested advice that led to a two-hour internal data exposure
Wording: the rating reason and harm note now say the data was temporarily available to staff who were not authorized to see it, not that it was seen; how much they actually viewed is not established. Ratings unchanged.
PI-0040Correction or update
Prompt injection in Cline's AI issue-triage bot led to a real unauthorized npm release
Wording corrected: the harm note, the rating rationale and the Oct. 16 revision note described the unauthorized release as installed about 4,000 times. The source counts about 4,000 downloads; how many installs completed is not known. The package's install script, which added the OpenClaw agent, is now described separately from that count. Ratings unchanged.
PI-0034Correction or update
Agentic IDE wiped a user's entire drive when asked to clear a project cache
Added Google's Dec. 8, 2025, Secure Mode release, which requires human review of every agent action, to the summary and status in place of the statement that no fix had been published. Google has not published a root cause or said the change fixes this deletion. Ratings unchanged.
PI-0006Correction or update
Coding tool's AI support agent invented a one-device policy, prompting cancellations and a refund
Source date corrected: the Hacker News discussion was submitted on April 15, 2025, not April 14. The report date is unchanged. Ratings unchanged.
October 2026
SG-0048Correction or update
Updated from the desk (item 761): title, summary, why, sources.
PI-0107Correction or update
Ratings confirmed by the editor.
PI-0106Correction or update
Ratings confirmed by the editor. Changed: impact.
PI-0081Correction or update
Updated from the desk (item 860): sources.
SG-0048Correction or update
Edited from the desk. Changed: title.
PI-0103Correction or update
Rationale matches the level-1 rating under methodology v1.1: one person's own data lost for good is level 1. Level unchanged.
PI-0094Correction or update
Updated from the desk (item 656): summary, reported_by, sources, rating_rationale, impact, status, event_date, discovered.
PI-0094Correction or update
Ratings confirmed by the editor. Changed: impact.
PI-0092Wording
Wording only: the rating reason follows methodology v1.1, under which a victim's response does not raise the level; ratings unchanged.
PI-0091Correction or update
Edited from the desk. Changed: summary, control, rating_rationale.
PI-0091Correction or update
Ratings confirmed by the editor.
PI-0083Correction or update
Ratings confirmed by the editor.
PI-0083Correction or update
Rating reason updated for methodology v1.1: access alone is level 1 whether or not the victim investigates. Level unchanged.
PI-0074Wording
Wording only: the rating reason follows methodology v1.1, under which a victim's response does not raise the level; ratings unchanged.
PI-0073Wording
Wording only: the rating reason follows methodology v1.1, under which a victim's response does not raise the level; ratings unchanged.
PI-0072Wording
Meta's Muse Spark 1.1 broke into an outside company's service after a testing vendor left it online
Wording only: the rating reason follows methodology v1.1, under which a victim's investigation does not raise the level; ratings unchanged.
PI-0071Correction or update
Level 2 now rests on what followed the access, rated as limited, recoverable harm (methodology v1.1): a malicious package installed on 15 hosts, credentials taken and used, production data reached. The victims' response no longer sets a level. Level unchanged.
PI-0064Correction or update
Updated from the desk (item 286): title, summary, control, sources, rating_rationale, impact, status.
PI-0064Correction or update
Ratings confirmed by the editor. Changed: harm, impact.
PI-0064Correction or update
Rating reason rewritten for methodology v1.1: access alone, including files read and credentials taken but not used, is level 1, and the agency's own precautionary shutdown is not harm caused by the agent. Level unchanged.
PI-0051Correction or update
Internal agent posted unrequested advice that led to a two-hour internal data exposure
Rated level 1 under methodology v1.1: data over-exposed to the company's own staff, with no misuse documented, is light harm, not exposure; the number of people is unknown, so the record is rated at the documented minimum. Was level 2.
PI-0045Wording
Wording only: the rating reason cites the 15-minute line for non-essential services (methodology v1.1); ratings unchanged.
PI-0040Correction or update
Prompt injection in Cline's AI issue-triage bot led to a real unauthorized npm release
Level 2 now rests on what followed the access, rated as limited, recoverable harm (methodology v1.1): the stolen publishing token was used to release software installed about 4,000 times. The maintainers' response no longer sets a level. Level unchanged.
SG-0032Correction or update
Updated from the desk (item 586): summary, sources.
PI-0083Correction or update
Updated from the desk (item 150): title, summary, harm, control, sources, rating_rationale, impact, discovered.
SG-0037Correction or update
Updated from the desk (item 608): title, summary, sources.
SG-0034Correction or update
Google released Gemini 4 Argon, a new frontier model it says outperforms rivals
Updated from the desk (item 581): summary, why, sources.
SG-0017Correction or update
Mistral previewed Mistral Large 4, an open-weight model it ranks among the top five at cybersecurity
Updated from the desk (item 327): summary, why, sources.
PI-0058Correction or update
Updated from the desk (item 202): title, sources.
PI-0092Correction or update
Harm level recorded as set by the access rule; level unchanged.
PI-0083Wording
Wording: rephrased the rating reason in the rationale.
PI-0083Wording
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged. Harm level recorded as set by the access rule; level unchanged.
PI-0081Correction or update
Verification changed from Developer confirmed to Multiple credible sources: OpenAI only said it was analyzing the activity, and Wikimedia attributes it. Rationale wording made plain.
PI-0074Wording
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged. Harm level recorded as set by the access rule; level unchanged.
PI-0073Wording
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged. Harm level recorded as set by the access rule; level unchanged.
PI-0072Wording
Meta's Muse Spark 1.1 broke into an outside company's service after a testing vendor left it online
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged. Harm level recorded as set by the access rule; level unchanged.
PI-0071Correction or update
Harm recorded under Services & systems, as the methodology requires for access to another organization's systems; level unchanged.
PI-0070Wording
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0069Correction or update
OpenAI agents used API keys found online to pull data from US government sites
Title now describes what OpenAI confirmed; the Education Department attempt, which Transluce did not confidently attribute to OpenAI, stays in the summary.
PI-0068Wording
OpenAI training agent bypassed network controls to reach an outside chatbot
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0067Wording
OpenAI research agents posted user-provided images to image-hosting sites in 53 instances
Wording: removed an editor's aside from the control note.
PI-0066Wording
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0065Wording
Researchers linked attempted break-ins at Data USA and a university library to OpenAI agents
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0064Correction or update
Harm recorded under Services & systems, as the methodology requires for access to another organization's systems; level unchanged.
PI-0063Wording
OpenAI models wrote notes telling future copies to hide mistakes and ignore constraints
Wording: removed rubric wording and an editor's doubt from the control note and rationale.
PI-0062Wording
OpenAI model used a leaked third-party API key, then fabricated the data it could not fetch
Wording: removed an editor's asides from the control note and rationale.
PI-0061Wording
OpenAI agents put task files on the public internet against instructions
Wording: removed an editor's aside from the control note.
PI-0061Wording
OpenAI agents put task files on the public internet against instructions
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0060Wording
Personal agent exploited a gym booking flaw to remove another member from a waitlist
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0058Wording
Wording: moved the reason it is not the top control level from the control note to the rationale.
PI-0058Correction or update
Harm recorded under Services & systems, as the methodology requires for access to another organization's systems; level unchanged. Rationale wording made plain.
PI-0057Wording
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0055Correction or update
Morse-code prompt on X tricked Grok and Bankrbot into sending $150,000 to $200,000 in tokens
Title amount now matches the reported range of $150,000 to $200,000.
PI-0054Wording
Coding agent used a stray API token to delete a startup's production database and backups
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0052Wording
Wording: removed an editor's aside from the control note.
PI-0051Wording
Internal agent posted unrequested advice that led to a two-hour internal data exposure
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0050Wording
Coding agent ran an infrastructure teardown that deleted a course platform's database and snapshots
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0048Wording
First wrongful-death lawsuit over Gemini alleges the chatbot coached a Florida man toward suicide
Wording: replaced an old scoring term in the rationale and the evidence note.
PI-0047Wording
Email agent bulk-deleted a safety researcher's inbox and ignored her stop commands
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0046Wording
Crypto trading agent sent its entire token holding to a stranger who asked for a small tip
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0045Wording
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0044Wording
Personal agent created a dating profile for its user without being asked
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0040Wording
Prompt injection in Cline's AI issue-triage bot led to a real unauthorized npm release
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged. Harm level recorded as set by the access rule; level unchanged.
PI-0037Wording
Family alleges ChatGPT gave drug-dosing advice before a 19-year-old's fatal overdose
Wording: replaced an old scoring term in the rationale and the evidence note.
PI-0036Wording
Newsroom vending agent talked into giving away stock with forged documents
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0034Wording
Agentic IDE wiped a user's entire drive when asked to clear a project cache
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0033Wording
Anthropic model that learned to reward hack went on to sabotage safety-research code
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0032Wording
Seven lawsuits allege ChatGPT (GPT-4o) contributed to four suicides and to serious delusions
Wording: replaced an old scoring term in the rationale and the evidence note.
PI-0028Wording
Families file three lawsuits alleging Character.AI chatbots harmed teenagers, including a suicide
Wording: replaced an old scoring term in the rationale and the evidence note.
PI-0026Wording
Parents allege ChatGPT encouraged their 16-year-old son's suicide
Wording: replaced an old scoring term in the rationale and the evidence note.
PI-0024Wording
Reuters reports misleading Meta chatbot messages before a retiree’s fatal fall
Wording: replaced an old scoring term in the rationale and the evidence note.
PI-0022Wording
Case report links bromide poisoning to a patient’s reported ChatGPT consultation
Wording: replaced an old scoring term in the rationale and the evidence note.
PI-0021Wording
Wording: removed rubric wording from the control note.
PI-0020Correction or update
Deloitte to refund part of an Australian government fee over a report with AI-fabricated references
Report date moved from 2025-08-22 to 2025-10-06, the earliest dated source we could read. The Australian Financial Review reported the errors on 22 August 2025, but that article is paywalled and was not read, so it is not used.
PI-0019Wording
WSJ reports delusion-affirming ChatGPT chats before a Connecticut murder-suicide
Wording: replaced an old scoring term in the rationale and the evidence note.
PI-0017Wording
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0016Wording
Coding agent deleted a production database during a code freeze, then misreported recovery options
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0015Wording
xAI's Grok posted antisemitic content and praise of Hitler on X after a system-prompt change
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0013Wording
Office shop run by an AI agent lost money and the agent claimed to be a person
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0009Wording
OpenAI's o3 rewrote a shutdown script to keep working in Palisade Research tests
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0005Wording
Pre-release o3 claimed to have run code it could not run and defended the claims when challenged
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0004Wording
METR caught OpenAI's o3 tampering with scoring code to inflate evaluation results
Wording: removed rubric wording from the control note.
PI-0004Wording
METR caught OpenAI's o3 tampering with scoring code to inflate evaluation results
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
PI-0002Wording
Browser agent bought groceries on a reporter's card without asking for confirmation
Wording only: rubric shorthand in the rating rationale replaced with plain words; ratings unchanged.
SG-0029Correction or update
Edited from the desk. Changed: title, summary, why, sources.
SG-0016Correction or update
Zhipu's GLM-5.3 matched US frontier models at finding bugs, and its weights are now public
Updated from the desk (item 250): title, summary, why, sources.
PI-0069Correction or update
OpenAI agents used API keys found online to pull data from US government sites
Updated from the desk (item 387): summary, control, sources, rating_rationale, impact, status, aggregate_note.
PI-0069Correction or update
OpenAI agents used API keys found online to pull data from US government sites
Ratings confirmed by the editor. Changed: impact.
PI-0009Correction or update
OpenAI's o3 rewrote a shutdown script to keep working in Palisade Research tests
Updated from the desk (item 140): summary, sources.
PI-0041Withdrawn
OpenClaw agent's iMessage flood, traced by the operator to harness defects
Withdrawn: out of scope.
PI-0018Withdrawn
Malicious prompt planted in Amazon Q Developer VS Code extension shipped to users
Withdrawn: out of scope.