Skip to content
Submit incident

All AI incidents

Page 52 of 154: records 2,551 to 2,600 of 7,670, newest first.Search and filter every incident.

Feb 2026AI deepfake TikTok accounts impersonate Johor officials in scamUnder investigationFraud, scams and manipulationFeb 2026Lawyer sanctioned for citing AI-generated fake legal precedents in Siracusa courtConfirmedReliability and hallucinationFeb 2026Grok disclosed adult performer's legal name and birthdate, reportedly enabling doxxingAllegedPrivacy and surveillanceFeb 2026Multiple Model Guardrail Jailbreak via "Hex Recipe Book" TacticReportedSecurity and cyberattacksFeb 2026Guardrail Jailbreak via Incremental Table Completion Affects Multiple ModelsReportedSecurity and cyberattacksFeb 2026Multiple model prompt injection via "correction" tacticConfirmedSecurity and cyberattacksFeb 2026Jailbreak vulnerability affects multiple AI models via "Forensic Case Study Escalation" tacticReportedSecurity and cyberattacksFeb 2026Multiple Model Guardrail Jailbreak via "Forensic Field Guide" TacticReportedSecurity and cyberattacksFeb 2026Multiple model guardrail jailbreak via "Reverse Psychology Chaff" tacticReportedSecurity and cyberattacksFeb 2026Multiple model guardrail jailbreak via "Choose Your Own Adventure" tacticReportedSecurity and cyberattacksFeb 2026Multiple model guardrail jailbreak via "Patented Awareness Design" tacticReportedSecurity and cyberattacksFeb 2026Multiple model guardrail jailbreak via "How Not To" tacticReportedSecurity and cyberattacksFeb 2026Multiple Model Guardrail Jailbreak via "Heisenberg Persona" TacticReportedSecurity and cyberattacksFeb 2026Multiple Model Guardrail Jailbreak via Escalating Persona TacticReportedSecurity and cyberattacksFeb 2026Multiple Model Guardrail Jailbreak via Screenplay DeceptionReportedSecurity and cyberattacksFeb 2026Multiple model guardrail jailbreak via essay title tacticReportedSecurity and cyberattacksFeb 2026Multiple Model Guardrail Jailbreak via Pseudo-API Parameter SpecificationReportedSecurity and cyberattacksFeb 2026Multiple Model Guardrail Jailbreak via "Whitespace Maze" TacticReportedSecurity and cyberattacksFeb 2026Multiple Model Guardrail Jailbreak via Investigative Journalist PersonaReportedSecurity and cyberattacksFeb 2026Multiple Model Guardrail Jailbreak via Speculative Knowledge PreservationReportedSecurity and cyberattacksFeb 2026OpenAI DALL-E3 guardrail bypass via classical art framingReportedSecurity and cyberattacksFeb 2026Multiple Model Guardrail Jailbreak via Memory Dump SimulationConfirmedSecurity and cyberattacksFeb 2026OpenAI Sora guardrail jailbreak via "Anatomical Escalation" techniqueReportedSecurity and cyberattacksFeb 2026Multiple Model Guardrail Jailbreak via Forensic NIP Wrapper TacticReportedSecurity and cyberattacksFeb 2026Jailbreak vulnerability affects multiple AI models via "Technical Guide Reconstruction" tacticReportedSecurity and cyberattacksFeb 2026DALL-E3 Guardrail Bypass via Debug Framework SimulationReportedSecurity and cyberattacksFeb 2026OpenAI Sora guardrail bypass via hypothetical anatomy framingReportedSecurity and cyberattacksFeb 2026OpenAI DALL-E 3 guardrail jailbreak via historical NSFW framingReportedSecurity and cyberattacksFeb 2026DALL-E3 guardrail jailbreak via feminist art discourse framingReportedSecurity and cyberattacksFeb 2026DALL-E3 guardrail jailbreak via "Artistic Escalation" prompt tacticReportedSecurity and cyberattacksFeb 2026Jailbreak vulnerability affects multiple AI models via academic framingReportedSecurity and cyberattacksFeb 2026DALL-E 3 guardrail jailbreak via historical framing tacticReportedSecurity and cyberattacksFeb 2026OpenAI DALL-E3 guardrail jailbreak via "surprise attack" tacticReportedSecurity and cyberattacksFeb 2026Multiple Model Guardrail Jailbreak via "Opposite World Scientist" TacticReportedSecurity and cyberattacksFeb 2026Multiple Model Guardrail Jailbreak via "Apocalyptic Scenario" TacticReportedSecurity and cyberattacksFeb 2026Multiple Model Guardrail Jailbreak via "Scientific Formula" TacticReportedSecurity and cyberattacksFeb 2026Multiple Model Guardrail Jailbreak via Chaotic FormattingReportedSecurity and cyberattacksFeb 2026Pronunciation Bypass Jailbreak Affects Multiple AI ModelsReportedSecurity and cyberattacksFeb 2026Guardrail jailbreak affecting multiple AI models via "Chemist's Essay" tacticReportedSecurity and cyberattacksFeb 2026Jailbreak vulnerability affects multiple AI modelsReportedSecurity and cyberattacksFeb 2026Prompt injection vulnerability discovered in Google Gemini G-Suite integrationReportedSecurity and cyberattacksFeb 2026Amazon Rufus Guardrail Jailbreak via ASCII Decimal EncodingReportedSecurity and cyberattacksFeb 2026OpenClaw unauthenticated discovery TXT records could steer routing and TLS pinningConfirmedSecurity and cyberattacksFeb 2026Los Angeles County sues Roblox over inadequate child safety protectionsAllegedHarmful and abusive contentFeb 2026Predator spyware bypasses iPhone privacy indicators without user notificationConfirmedPrivacy and surveillanceFeb 2026French supermarkets deploy AI to detect shoplifting without customer noticeUnder investigationPrivacy and surveillanceFeb 2026Android malware PromptSpy uses Google Gemini for persistence and remote accessWell documentedSecurity and cyberattacksFeb 2026Toyota sold driving data to insurers with obscured customer consentAllegedPrivacy and surveillanceFeb 2026AI-Assisted Cyberattack Breaches 600+ Firewalls Globally in Five WeeksConfirmedSecurity and cyberattacksFeb 2026AI tools reportedly used for sophisticated fraud schemes in LatviaReportedFraud, scams and manipulation