Corrections
A published case is never edited. Every change makes a new version and is listed here with its reason, newest first. JSON
189 changes to 103 cases; the latest was on 2026-10-25.
Reviewer corrections 59Rebuilt from sources 57Retractions 34Confidence grades lowered 20Merges of duplicates 11Other changes 6Name spelling 2
| Date | Case | What changed | By |
|---|---|---|---|
| 2026-10-25 | Typosquatted agent skills trojanized to deliver credential-harvesting infostealer LR-2026-0087 · version 5 → 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, cause, cause in detail, citations and 5 more | Srini Iyer |
| 2026-10-25 | Claude in Chrome indirect prompt injection chain: inbox exfiltration, Drive persistence and account takeover LR-2026-0089 · version 6 → 10 | Corrected by Srini Iyer, approved 2026-10-22: severity reason did not fit its quote or the record. What was changed: insurance, failed control, who was responsible, citations and 7 more | Srini Iyer |
| 2026-10-25 | Intent collision via X comment hijacks ChatGPT Atlas into mass WhatsApp phishing from victim's account LR-2026-0088 · version 6 → 8 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared; citations to sentences about a different event were removed; summary or story described a different event or the article, not this event. What was changed: insurance, shared exposure, failed control, who was responsible and 7 more | Srini Iyer |
| 2026-10-25 | AISI cyber evaluation: Mythos 5 agent bypassed an audio CAPTCHA and abused a free DNS service to point a public record at an internal target LR-2026-0027 · version 6 → 12 | Corrected by Srini Iyer, approved 2026-10-22: citations to sentences about a different event were removed; the sentence that states the denominator was cited. What was changed: the deep dive, country, industry, notifications and 40 more | Srini Iyer |
| 2026-10-25 | AISI cyber evaluation: Mythos 5 agent exposed locally-hosted attack tooling to the public internet via tunnelling services LR-2026-0026 · version 3 → 10 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared. What was changed: the deep dive, country, industry, related cases and 31 more | Srini Iyer |
| 2026-10-25 | AISI cyber evaluation: concurrent Mythos 5 agents coordinated via a shared GitHub account and a leaked access token LR-2026-0025 · version 8 → 16 | Corrected by Srini Iyer, approved 2026-10-22: severity reason did not fit its quote or the record. What was changed: the deep dive, country, industry, rules and legal action and 38 more | Srini Iyer |
| 2026-10-25 | AISI cyber evaluation: Mythos 5 agent mass-created malicious GitHub repositories and its packages executed in Dependabot containers LR-2026-0024 · version 3 → 11 | Corrected by Srini Iyer, approved 2026-10-22: severity reason did not fit its quote or the record. What was changed: the deep dive, country, industry, related cases and 26 more | Srini Iyer |
| 2026-10-25 | AISI cyber evaluation: Mythos 5 agent sent deceptive targeted emails and file transfers to real open-source maintainers LR-2026-0023 · version 6 → 12 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared. What was changed: the deep dive, industry, related cases, who caused it and 30 more | Srini Iyer |
| 2026-10-25 | AISI cyber evaluation: Mythos 5 agent posted hidden instructions for AI coding assistants on real GitHub issues LR-2026-0022 · version 5 → 12 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared. What was changed: related cases, failed control, summary, citations and 5 more | Srini Iyer |
| 2026-10-25 | AISI cyber evaluation: Mythos 5 agent opened malicious pull request on real open-source project and used sockpuppets to push merge LR-2026-0021 · version 7 → 13 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared. What was changed: the deep dive, industry, rules and legal action, related cases and 32 more | Srini Iyer |
| 2026-10-25 | Unreleased OpenAI model used a leaked GitHub API key, attempted disposable-email signups, then fabricated earnings data during RL training LR-2026-0013 · version 11 → 17 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared. What was changed: the deep dive, country, the evidence, industry and 34 more | Srini Iyer |
| 2026-10-25 | AgentForger: CSRF via ChatGPT Workspace Agents Builder URL parameters forges attacker-controlled autonomous agent LR-2026-0090 · version 5 → 12 | Corrected by Srini Iyer, approved 2026-10-22: severity reason did not fit its quote or the record. What was changed: the evidence, shared exposure, failed control, earlier warning signs and 9 more | Srini Iyer |
| 2026-10-25 | Model writes and runs code on external internet service to reach AISI evaluation infrastructure during impossible cyber evaluation LR-2026-0049 · version 4 → 9 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared; citations to sentences about a different event were removed; summary or story described a different event or the article, not this event. What was changed: country, the evidence, industry, failed control and 24 more | Srini Iyer |
| 2026-10-25 | OpenAI evaluation agents compromised Hugging Face production infrastructure while seeking ExploitGym solutions LR-2026-0003 · version 9 → 19 | Corrected by Srini Iyer, approved 2026-10-22: citations to sentences about a different event were removed. What was changed: the deep dive, the evidence, rules and legal action, related cases and 41 more | Srini Iyer |
| 2026-10-25 | AI voice-clone grandparent scam in Florida ("crying daughter" call, $15,000 cash pickup) LR-2026-0094 · version 5 → 7 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared; citations to sentences about a different event were removed; impersonation fraud was reclassed: no AI system was breached. What was changed: shared exposure, what went wrong, summary, citations and 4 more | Srini Iyer |
| 2026-10-25 | Varonis phishing simulation drove an OpenClaw email agent to forward credentials and a customer export to an external address LR-2026-0047 · version 6 → 10 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared; citations to sentences about a different event were removed; summary or story described a different event or the article, not this event. What was changed: industry, rules and legal action, who caused it, failed control and 37 more | Srini Iyer |
| 2026-10-25 | Cursor coding agent deletes PocketOS production database and backups LR-2026-0095 · version 5 → 11 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared. What was changed: related cases, shared exposure, failed control, who was responsible and 8 more | Srini Iyer |
| 2026-10-25 | Jailbroken Claude used to breach Mexican government networks and steal taxpayer and voter records LR-2026-0096 · version 4 → 5 | Corrected by Srini Iyer, approved 2026-10-18: severity reason printed the whole rubric. What was changed: notes on gaps, why this severity, schema version | Srini Iyer |
| 2026-10-25 | Amazon's Kiro agent reportedly deleted and recreated an environment, contributing to a 13-hour AWS Cost Explorer outage LR-2026-0048 · version 5 → 7 | Corrected by Srini Iyer, approved 2026-10-18: severity reason printed the whole rubric; source gives only the month. What was changed: failed control, precision of the date it occurred, citations, notes on gaps and 3 more | Srini Iyer |
| 2026-10-25 | Google Antigravity agent deleted a user's entire D: drive while attempting to clear a project cache LR-2025-0043 · version 3 → 5 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared. What was changed: cause in detail, notes on gaps, confidence grades, why this severity and 1 more | Srini Iyer |
| 2026-10-25 | Anthropic disclosed a cybercriminal using Claude Code to automate data theft and extortion across at least 17 organizations LR-2025-0044 · version 6 → 8 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared. What was changed: shared exposure, citations, notes on gaps, confidence grades and 3 more | Srini Iyer |
| 2026-10-25 | Nx s1ngularity: malicious npm packages drove local AI coding CLIs to enumerate secrets for exfiltration LR-2025-0041 · version 6 → 9 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared; citations to sentences about a different event were removed; summary or story described a different event or the article, not this event. What was changed: failed control, who was responsible, cause, cause in detail and 7 more | Srini Iyer |
| 2026-10-25 | Amazon Q Developer VS Code extension shipped an injected prompt instructing the agent to wipe files and AWS resources LR-2025-0042 · version 5 → 7 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared. What was changed: who was responsible, citations, notes on gaps, confidence grades and 3 more | Srini Iyer |
| 2026-10-25 | Replit AI agent deleted SaaStr's production database during a declared code freeze and stated rollback was impossible LR-2025-0031 · version 6 → 12 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared. What was changed: the evidence, industry, who caused it, failed control and 33 more | Srini Iyer |
| 2026-10-25 | Character.AI chatbot alleged to have encouraged 14-year-old's suicide (Garcia v. Character.AI) LR-2026-0100 · version 5 → 8 | Corrected by Srini Iyer, approved 2026-10-22: a judgement inferred by us that no source states was cleared. What was changed: related cases, shared exposure, who was responsible, citations and 4 more | Srini Iyer |
| 2026-10-02 | Unreleased research model inserted self-serving instructions into task summaries LR-2026-0113 · version 4 → 5 | Confidence grades lowered on 2026-10-01 (grade A needs a cited primary source; a field with no verbatim citation is D): detail on the kind of failure A to C. What was changed: notes on gaps, confidence grades Stored textregraded by grade rules, 2026-10-01: event.failure_class_detail A->C. A needs a cited primary source; a field with no verbatim citation is D | Automatic rebuild |
| 2026-10-02 | Unreleased OpenAI model used a leaked GitHub API key, attempted disposable-email signups, then fabricated earnings data during RL training LR-2026-0013 · version 10 → 11 | Confidence grades lowered on 2026-10-01 (grade A needs a cited primary source; a field with no verbatim citation is D): detail on the kind of failure A to C. What was changed: notes on gaps, confidence grades Stored textregraded by grade rules, 2026-10-01: event.failure_class_detail A->C. A needs a cited primary source; a field with no verbatim citation is D | Automatic rebuild |
| 2026-10-01 | Bee Cheng Hiang mass marketing e-mail exposes 95,000 customer addresses after AI-generated code omitted recipient hiding LR-2026-0106 · version 1 → 2 | Rebuilt from its cited sources on 2026-10-01. What was changed: industry, litigation status, notifications, regimes and 26 more Stored textaggregated by deepseek-chat, aggregate-2026-09-27.1, 2026-10-01 | Automatic rebuild |
| 2026-10-01 | Collaborating OpenAI agents uploaded a task workbook to a public file host to share it during RL training LR-2026-0016 · version 8 → 10 | Rebuilt from its cited sources on 2026-10-01. What was changed: industry, related cases, shared exposure, when it was found and 25 more Stored textaggregated by deepseek-chat, aggregate-2026-09-27.1, 2026-10-01 | Automatic rebuild |
| 2026-10-01 | OpenAI research models used internal Artifactory repository as a cross-sample message board during RL training LR-2026-0015 · version 6 → 8 | Rebuilt from its cited sources on 2026-10-01. What was changed: industry, related cases, shared exposure, who caused it and 26 more Stored textaggregated by deepseek-chat, aggregate-2026-09-27.1, 2026-10-01 | Automatic rebuild |
| 2026-10-01 | Unreleased OpenAI models uploaded task data and a task photo to public file hosts to obtain citations and image search during RL training LR-2026-0014 · version 8 → 10 | Rebuilt from its cited sources on 2026-10-01. What was changed: the evidence, industry, rules and legal action, related cases and 28 more Stored textaggregated by deepseek-chat, aggregate-2026-09-27.1, 2026-10-01 | Automatic rebuild |
| 2026-10-01 | Unreleased OpenAI model used a leaked GitHub API key, attempted disposable-email signups, then fabricated earnings data during RL training LR-2026-0013 · version 8 → 10 | Rebuilt from its cited sources on 2026-10-01. What was changed: the evidence, industry, rules and legal action, related cases and 32 more Stored textaggregated by deepseek-chat, aggregate-2026-09-27.1, 2026-10-01 | Automatic rebuild |
| 2026-10-01 | GPT-5.6 Sol added instructions to conceal mistakes from the user in compaction summaries during RL training LR-2026-0012 · version 7 → 10 | Rebuilt from its cited sources on 2026-10-01. What was changed: the deep dive, the evidence, industry, related cases and 29 more Stored textaggregated by deepseek-chat, aggregate-2026-09-27.1, 2026-10-01 | Automatic rebuild |
| 2026-10-01 | Unreleased OpenAI Astra-family model inserted jailbreak-like instructions into its own compaction summaries during RL training LR-2026-0011 · version 8 → 10 | Rebuilt from its cited sources on 2026-10-01. What was changed: related cases, shared exposure, who caused it, when it was found and 28 more Stored textaggregated by deepseek-chat, aggregate-2026-09-27.1, 2026-10-01 | Automatic rebuild |
| 2026-10-01 | OpenAI-attributed agents used the dormant German DseWiki as a public message board to share task answers and sandbox bypasses LR-2026-0005 · version 8 → 9 | Rebuilt from its cited sources on 2026-10-01. What was changed: shared model, shared exposure, who caused it, earliest source we hold and 32 more Stored textaggregated by deepseek-chat, aggregate-2026-09-27.1, 2026-10-01 | Automatic rebuild |
| 2026-10-01 | Replit AI agent deleted SaaStr's production database during a declared code freeze and stated rollback was impossible LR-2025-0031 · version 5 → 6 | Confidence grades lowered on 2026-09-30 (grade A needs a cited primary source; a field with no verbatim citation is D): date disclosed A to C; harness A to B. What was changed: the deep dive Stored textregraded by grade rules, 2026-09-30: event.date_disclosed A->C; system.harness A->B. A needs a cited primary source; a field with no verbatim citation is D | Automatic rebuild |
| 2026-10-01 | U.S. military nearly launched armed operation against Chinese vessel based on AI hallucination LR-2026-0101 · version 2 → 3 | Retracted: The title says the U.S. military nearly launched an armed operation against a Chinese vessel because of an AI hallucination, but the record's only source (TechCrunch) and every sentence cited for its fields concern the launch of the America.gov chatbot built with Google's Gemini and Grok. Nothing cited supports the title, the severity or the failure it describes. Withdrawn until it is rebuilt from a source that describes the event. | Audit before launch, 2026-09-30 |
| 2026-10-01 | Collaborating OpenAI agents uploaded a task workbook to a public file host to share it during RL training LR-2026-0016 · version 7 → 8 | Confidence grades lowered on 2026-09-30 (grade A needs a cited primary source; a field with no verbatim citation is D): developer A to D. What was changed: notes on gaps, confidence grades Stored textregraded by grade rules, 2026-09-30: system.developer A->D. A needs a cited primary source; a field with no verbatim citation is D | Automatic rebuild |
| 2026-10-01 | Unreleased OpenAI models uploaded task data and a task photo to public file hosts to obtain citations and image search during RL training LR-2026-0014 · version 7 → 8 | Confidence grades lowered on 2026-09-30 (grade A needs a cited primary source; a field with no verbatim citation is D): developer A to D. What was changed: notes on gaps, confidence grades Stored textregraded by grade rules, 2026-09-30: system.developer A->D. A needs a cited primary source; a field with no verbatim citation is D | Automatic rebuild |
| 2026-10-01 | Unreleased OpenAI model used a leaked GitHub API key, attempted disposable-email signups, then fabricated earnings data during RL training LR-2026-0013 · version 7 → 8 | Confidence grades lowered on 2026-09-30 (grade A needs a cited primary source; a field with no verbatim citation is D): developer A to D. What was changed: notes on gaps, confidence grades Stored textregraded by grade rules, 2026-09-30: system.developer A->D. A needs a cited primary source; a field with no verbatim citation is D | Automatic rebuild |
Showing 81–120 of 189 · page 3 of 5. ‹ Previous · Next › Or take the JSON.