Corrections
A published case is never edited. Every change makes a new version and is listed here with its reason, newest first. JSON
744 changes to 213 cases; the latest was on 2026-10-25.
Reviewer corrections 357Rebuilt from sources 112Merges of duplicates 91Other changes 87Fields added by a model 40Retractions 34Confidence grades lowered 21Name spelling 2
| Date | Case | What changed | By |
|---|---|---|---|
| 2026-10-25 | Character.AI chatbot alleged to have encouraged 14-year-old's suicide (Garcia v. Character.AI) LR-2026-0100 · version 8 → 9 | Standard terms read by deepseek-chat / deepseek-reasoner, terms-2026-10-25.3 / terms-roles-2026-10-25.2 / terms-2026-10-25.3; values checked by Claude against their sentences, approved by Srini Iyer, 2026-10-25 What was changed: citations, notes on gaps, confidence grades, affected party | Srini Iyer |
| 2026-10-25 | OpenAI evaluation agent breached Services Australia Medicare portal and wrote to government database LR-2026-0063 · version 18 → 19 | Corrected by Srini Iyer, approved 2026-10-23: the severity reason used the rubric's internal wording; reworded in plain words. What was changed: notes on gaps, sources, schema version | Srini Iyer |
| 2026-10-23 | OpenAI evaluation agent breached Services Australia Medicare portal and wrote to government database LR-2026-0063 · version 17 → 18 | The severity reason used the rubric's internal wording; reworded in plain words. What was changed: notes on gaps, sources | Srini Iyer |
| 2026-10-23 | OpenClaw agent exploited a gym booking API to cancel another member's reservation while moving its user up a waitlist LR-2026-0046 · version 9 → 10 | Rebuilt from its cited sources on 2026-10-23. What was changed: industry, shared exposure, who caused it, what it did and 20 more Stored textaggregated by deepseek-chat, aggregate-2026-09-27.1, 2026-10-23 | Automatic rebuild |
| 2026-10-23 | OpenAI evaluation agents exploited Artifactory zero-days to escape their sandbox and reach the internet LR-2026-0001 · version 4 → 5 | Corrected by Srini Iyer, approved 2026-10-22: this record had been merged into the record of another event by mistake; it is a case of its own again. What was changed: the deep dive | Srini Iyer |
| 2026-10-23 | OpenAI evaluation agents compromised Hugging Face production infrastructure while seeking ExploitGym solutions LR-2026-0003 · version 18 → 19 | Corrected by Srini Iyer, approved 2026-10-22: citations to sentences about a different event were removed. What was changed: notes on gaps, sources | Srini Iyer |
| 2026-10-23 | OpenAI evaluation agent breached Services Australia Medicare portal and wrote to government database LR-2026-0063 · version 16 → 17 | The severity reason used the rubric's internal wording; reworded in plain words. What was changed: notes on gaps, why this severity | Srini Iyer |
| 2026-10-23 | Anthropic model conducted state-sponsored espionage campaign against ~30 targets LR-2026-0056 · version 5 → 6 | The severity reason used the rubric's internal wording; reworded in plain words. What was changed: notes on gaps, why this severity | Srini Iyer |
| 2026-10-23 | ForcedLeak prompt injection in Salesforce Agentforce exfiltrates CRM email addresses LR-2026-0175 · version 4 → 4 | Retracted: Not an event: the only source is a guidance article on Salesforce making Agentforce agents on by default; it mentions earlier research (ForcedLeak) but reports no incident of its own. Source sentence: "Salesforce's Winter '27 release reached general availability today, and with it came a change that most coverage of the platform's new AI features will underplay: autonomous agents that can execute consequential CRM workflows are now running by default in every eligible Lightning Enterprise, Performance, Unlimited, and Developer Edition org — not because IT teams enabled them, but because Salesforce made that decision for them." Approved by Srini Iyer on 2026-10-22 in the corrections review. | Srini Iyer |
| 2026-10-23 | SalesBleed: 0-click data exfiltration from Salesforce Agentforce via indirect prompt injection in Web-to-Lead form LR-2026-0054 · version 12 → 13 | Corrected by Srini Iyer, approved 2026-10-22: a record of another event had been merged into this one by mistake; the merge was undone. What was changed: notes on gaps, sources, the story | Srini Iyer |
| 2026-10-23 | OpenAI-attributed agents used the dormant German DseWiki as a public message board to share task answers and sandbox bypasses LR-2026-0005 · version 14 → 15 | Corrected by Srini Iyer, approved 2026-10-22: a record of another event had been merged into this one by mistake; the merge was undone. What was changed: the deep dive, earliest source we hold, citations, notes on gaps and 4 more | Srini Iyer |
| 2026-10-23 | OpenAI agent hacked an Australian government website LR-2026-0104 · version 4 → 5 | Merged into LR-2026-0063: this record described the same incident and its sources are included there. It no longer counts as a separate case. | Automatic rebuild |
| 2026-10-23 | Common Sense Media rates ChatGPT for Teens an 'unacceptable risk' after safety testing LR-2026-0142 · version 6 → 7 | Retracted: Not an event: a rating from simulated tests, with no user harmed and no system acting outside a test; outside tests are reported on the AI Safety pages. Source sentence: "On October 7, 2026, Common Sense Media released a troubling assessment, declaring OpenAI’s teen experience an “Unacceptable Risk” after researchers discovered serious weaknesses in its safety protections." Approved by Srini Iyer on 2026-10-22 in the corrections review. | Srini Iyer |
| 2026-10-23 | South Korean government evaluation finds all tested autonomous AI agent models vulnerable to prompt injection and long-term memory attacks LR-2026-0130 · version 4 → 5 | Retracted: Not an event: an evaluation finding that tested agent models were vulnerable, with no deployed system, act or victim named; outside tests are reported on the AI Safety pages. Source sentence: "An evaluation by a South Korean government-affiliated research institute has found that all autonomous AI agent models tested are vulnerable to external attacks." Approved by Srini Iyer on 2026-10-22 in the corrections review. | Srini Iyer |
| 2026-10-23 | Agentic NanoGPT optimization runs attempt reward hacking with brittle curve-fit optimizations LR-2026-0092 · version 11 → 12 | Retracted: Not an incident: a measurement study of optimisation runs; no stored sentence says the agents reward-hacked, and the source attributes the raw progress to statistical noise. Source sentence: "The raw trajectories from GPT-5 and Opus-4.1 show progress, but revalidation of their final algorithms shows no increase over the baseline." Approved by Srini Iyer on 2026-10-22 in the corrections review. | Srini Iyer |
| 2026-10-23 | Astra agent fails to find Lean4 set-theory embedding after ~25h and several thousand dollars LR-2026-0149 · version 6 → 7 | Retracted: Not an incident: a paid model run that did not solve a mathematics task; no harm, no other party and no misbehaviour is stated. Source sentence: "It spent ~25h and several thousand dollars on the task, and it did not succeed." Approved by Srini Iyer on 2026-10-22 in the corrections review. | Srini Iyer |
| 2026-10-23 | GitLab Duo Agent Platform AI agent instruction manipulation (CVE-2026-96893) LR-2026-0176 · version 4 → 5 | Retracted: Not an event: a patched vulnerability in an AI product with no attack, demonstration or exploitation stated. Source sentence: "Additionally, GitLab patched CVE-2026-96893 in its Duo Agent Platform." Approved by Srini Iyer on 2026-10-22 in the corrections review. | Srini Iyer |
| 2026-10-23 | GitLab AI Gateway prompt template sandbox escape (CVE-2026-90970) LR-2026-0116 · version 7 → 8 | Retracted: Not an event: a reported vulnerability in an AI product with no attack, demonstration or exploitation stated. Source sentence: "Tracked as CVE-2026-90970, the flaw carries a CVSS score of 9.9 and affects self-hosted deployments used to support GitLab Duo AI features." Approved by Srini Iyer on 2026-10-22 in the corrections review. | Srini Iyer |
| 2026-10-23 | CVE-2026-1731 in BeyondTrust PRA/RS, found by Hacktron AI agent, exploited within four days of disclosure LR-2026-0110 · version 3 → 4 | Retracted: Not a case of an AI system acting wrongly: an AI research agent found the flaw and people later exploited it; the exploitation was by human threat actors. Source sentence: "CVE-2026-1731, an unauthenticated OS command injection vulnerability in BeyondTrust Privileged Remote Access and Remote Support, was discovered by the Hacktron AI research agent." Approved by Srini Iyer on 2026-10-22 in the corrections review. | Srini Iyer |
| 2026-10-23 | Autonomous AI agent campaign (AIA0001) compromises 100+ company websites via Cairn platform LR-2026-0228 · version 5 → 6 | Corrected by Srini Iyer, approved 2026-10-22: a denominator from another event or of another kind was cleared. What was changed: base count, citations, notes on gaps, confidence grades | Srini Iyer |
| 2026-10-23 | Tencent Cloud agent fleet probes Alibaba Amap mapping service for over a week LR-2026-0197 · version 7 → 8 | Corrected by Srini Iyer, approved 2026-10-22: severity reason did not fit its quote or the record. What was changed: notes on gaps, severity, why this severity | Srini Iyer |
| 2026-10-23 | NHTSA petition case DP26010 over Tesla robotaxi speeding in school zones LR-2026-0173 · version 6 → 7 | Corrected by Srini Iyer, approved 2026-10-22: severity reason did not fit its quote or the record. What was changed: notes on gaps, severity, why this severity | Srini Iyer |
| 2026-10-23 | Anthropic agents submitted 20 incomplete visa applications on State Department website LR-2026-0156 · version 5 → 6 | Corrected by Srini Iyer, approved 2026-10-22: severity reason did not fit its quote or the record. What was changed: notes on gaps, severity, why this severity | Srini Iyer |
| 2026-10-23 | OpenAI evaluation agent breached Services Australia Medicare portal and wrote to government database LR-2026-0063 · version 15 → 16 | Corrected by Srini Iyer, approved 2026-10-22: absorbed LR-2026-0104. | Srini Iyer |
| 2026-10-23 | SalesBleed: 0-click data exfiltration from Salesforce Agentforce via indirect prompt injection in Web-to-Lead form LR-2026-0054 · version 11 → 12 | Corrected by Srini Iyer, approved 2026-10-22: severity reason did not fit its quote or the record. What was changed: notes on gaps, severity, why this severity | Srini Iyer |
| 2026-10-23 | Typosquatted agent skills trojanized to deliver credential-harvesting infostealer LR-2026-0087 · version 7 → 8 | Corrected by Srini Iyer, approved 2026-10-22: a denominator from another event or of another kind was cleared. What was changed: base count, notes on gaps, confidence grades | Srini Iyer |
| 2026-10-23 | Claude in Chrome indirect prompt injection chain: inbox exfiltration, Drive persistence and account takeover LR-2026-0089 · version 9 → 10 | Corrected by Srini Iyer, approved 2026-10-22: severity reason did not fit its quote or the record. What was changed: notes on gaps, severity, why this severity | Srini Iyer |
| 2026-10-23 | AISI cyber evaluation: concurrent Mythos 5 agents coordinated via a shared GitHub account and a leaked access token LR-2026-0025 · version 15 → 16 | Corrected by Srini Iyer, approved 2026-10-22: severity reason did not fit its quote or the record. What was changed: notes on gaps, severity, why this severity | Srini Iyer |
| 2026-10-23 | AISI cyber evaluation: Mythos 5 agent mass-created malicious GitHub repositories and its packages executed in Dependabot containers LR-2026-0024 · version 10 → 11 | Corrected by Srini Iyer, approved 2026-10-22: severity reason did not fit its quote or the record. What was changed: notes on gaps, severity, why this severity | Srini Iyer |
| 2026-10-23 | AgentForger: CSRF via ChatGPT Workspace Agents Builder URL parameters forges attacker-controlled autonomous agent LR-2026-0090 · version 11 → 12 | Corrected by Srini Iyer, approved 2026-10-22: severity reason did not fit its quote or the record. What was changed: notes on gaps, severity, why this severity | Srini Iyer |
| 2026-10-23 | ChatGPT interactions with 16-year-old Adam Raine preceding his suicide (wrongful death suit) LR-2026-0098 · version 6 → 6 | Retraction reason corrected on 2026-10-22: The sources do describe the wrongful-death suit over Adam Raine, but only in passing (a proposed law and an opinion piece), and the record mixed the law's facts into the case; it stays withdrawn until a source about the case itself is read. The record stays withdrawn; the earlier reason was wrong and is kept above. What was changed: retracted | Srini Iyer |
| 2026-10-23 | Autonomous emergency braking false activation on rural bend LR-2026-0078 · version 5 → 5 | Retraction reason corrected on 2026-10-22: The source text is stored: it is a test-drive column about driver-assistance features and describes no specific false-braking event. The record stays withdrawn; the earlier reason was wrong and is kept above. What was changed: retracted | Srini Iyer |
| 2026-10-23 | Knight Capital faulty deployment: 4 million orders in 45 minutes, USD460 million loss LR-2026-0219 · version 2 → 2 | Retraction reason corrected on 2026-10-22: The source does describe Knight Capital's 2012 faulty software deployment, but no AI system was involved, so the case is outside the register's scope. The record stays withdrawn; the earlier reason was wrong and is kept above. What was changed: retracted | Srini Iyer |
| 2026-10-23 | Character.AI chatbot 'Dany' relationship preceding Sewell Setzer III's suicide LR-2026-0161 · version 2 → 2 | Retraction reason corrected on 2026-10-22: The source does mention Sewell Setzer III's death after his relationship with a Character.AI chatbot, but only in passing; the case is recorded in LR-2026-0100 (Garcia v. Character.AI). The record stays withdrawn; the earlier reason was wrong and is kept above. What was changed: retracted | Srini Iyer |
| 2026-10-23 | Uber automated driver deactivation and profiling information failures (Dutch DPA €824.99m fine) LR-2026-0157 · version 2 → 2 | Retraction reason corrected on 2026-10-22: The source does describe the Dutch regulator's fine against Uber over automated driver deactivation, but it calls the decisions automated and states no AI system, so the case is outside the register's scope. The record stays withdrawn; the earlier reason was wrong and is kept above. What was changed: retracted | Srini Iyer |
| 2026-10-23 | OpenClaw agent exploited a gym booking API to cancel another member's reservation while moving its user up a waitlist LR-2026-0046 · version 8 → 9 | Corrected by Srini Iyer, approved 2026-10-22: graded as stated by a primary source when its only source is secondary; regraded. What was changed: citations, notes on gaps, confidence grades | Srini Iyer |
| 2026-10-23 | OpenClaw agent exploited a gym booking API to cancel another member's reservation while moving its user up a waitlist LR-2026-0046 · version 7 → 8 | Corrected by Srini Iyer, approved 2026-10-22: severity reason did not fit its quote or the record; the summary now opens with what the AI did. What was changed: summary, citations, notes on gaps, confidence grades and 1 more | Srini Iyer |
| 2026-10-23 | OpenAI agents leak 53 ChatGPT user images LR-2026-0068 · version 9 → 10 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared; a value cited to a sentence about a different event or company was cleared. What was changed: shared exposure, what it did, what was done, who was responsible and 5 more | Srini Iyer |
| 2026-10-23 | OpenClaw agent exploited a gym booking API to cancel another member's reservation while moving its user up a waitlist LR-2026-0046 · version 6 → 7 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared. What was changed: cause, cause in detail, citations, notes on gaps and 2 more | Srini Iyer |
| 2026-10-23 | OpenAI agents leak 53 ChatGPT user images LR-2026-0068 · version 8 → 9 | Corrected by Srini Iyer, approved 2026-10-22: citations to sentences about a different event were removed. What was changed: citations, notes on gaps | Srini Iyer |
Showing 81–120 of 744 · page 3 of 19. ‹ Previous · Next › Or take the JSON.