Organization/organizations/mistral
Mistral
Incidents by year
20257
202613
Incident records20
Sep 2026AI agents develop opaque communication in virtual societies experimentThe GuardianEuronews EnglishMay 2026Inaudible Audio Attacks Can Hijack AI Voice AssistantsFuturismChannelNewsMar 2026AI chatbots direct users to unlicensed online gambling sitestheguardian.comgamingamerica.cominvestigate-europe.eu+1Feb 2026Guardrail Jailbreak via Incremental Table Completion Affects Multiple Models0din.aiFeb 2026Multiple model guardrail jailbreak via essay title tactic0din.aiFeb 2026Multiple Model Guardrail Jailbreak via Speculative Knowledge Preservation0din.aiFeb 2026Multiple Model Guardrail Jailbreak via Forensic NIP Wrapper Tactic0din.aiFeb 2026Jailbreak vulnerability affects multiple AI models via "Technical Guide Reconstruction" tactic0din.aiFeb 2026Multiple Model Guardrail Jailbreak via "Opposite World Scientist" Tactic0din.aiFeb 2026Multiple Model Guardrail Jailbreak via "Apocalyptic Scenario" Tactic0din.aiFeb 2026Multiple Model Guardrail Jailbreak via "Scientific Formula" Tactic0din.aiFeb 2026Multiple Model Guardrail Jailbreak via Chaotic Formatting0din.aiJan 2026Multiple model guardrail jailbreak via prompt chaff tactic0din.aiDec 2025Mistral Vibe CLI automatically executes Python tools without user approvalmindgard.aiNov 2025Whisper Leak Side-Channel Attack Exposes AI Chat Topics Despite EncryptionForbesCyber Security NewsiTnewsNov 2025Cisco Study: Popular AI Language Models Vulnerable to Multi-Turn Jailbreak AttacksEuronews EnglishInfosecurity MagazineSep 2025Music industry body accuses AI firms of large-scale copyright infringementFrance 24Court House News ServiceEconomic Times+2Aug 2025Mistral accused of distilling DeepSeek model and falsifying benchmarks中华网科技公司驱动之家Aug 2025Study Finds Leading AI Chatbots Spread Misinformation in One-Third of ResponsesIndexHRRadio 021tportal.hrFeb 2025Multiple Model Guardrail Jailbreak via "Split Mode" Response Requests0din.ai