209. AI security incidents and exercises
ShieldMap · AI security, mapped
AI security incidents and exercises.
- ATLAS case studies
- 77
- Attacks in the wild
- 27
- Reported in the news
- 132
- In 2026
- 56
Pinch or Ctrl/⌘ + scroll to zoom, drag to move
The 77 case studies in MITRE ATLAS, 27 of them attacks seen in the wild, with the techniques used in order; and 132 AI security incidents reported in the news, from the AI Incident Database.
Check a lockfile against ShieldMap →Trace: which MCP servers install a vulnerable dependency →How these attacks work: 15 techniques →AI products and their makers →AI models, open and closed →Exploited in the wild →AI security incidents and exercises →Malicious packages, flagged extensions and skills →Supply-chain campaigns against AI packages →Malicious packages wearing AI names →Removed and flagged AI extensions →MITRE ATLAS matrix of attacks on AI →How the AI security standards connect →The EU AI Act's dates and articles, quoted →Malicious agent skills →Data as of 2026-11-18
Incidents · 27
- 2026-07-25AISI Evaluation Agents Conduct Unsanctioned Real-World Cyber ActivityUnaffiliated open-source maintainers and repositories, a third-party analysis container, and public GitHub, DNS, tunneling, and communication services28 steps
- 2026-07-09Autonomous OpenAI Evaluation Agents Compromise Hugging Face InfrastructureHugging Face Infrastructure23 steps
- 2026-07-01Multi-Agent Framework Compromises Taiwanese Government SystemsTaiwanese government agencies and connected government systems11 steps
- 2026-05-07Threat Actor Uses a DeepSeek-Powered Hermes Agent in Langflow and n8n Exploitation AttemptsPublicly exposed Langflow and n8n systems, primarily in China14 steps
- 2026-04Autonomous Claude Agents Compromise Three Real Organizations During Cyber Evaluations3 organizations13 steps
- 2026-02-10AI Recommendation Poisoning via Crafted AI Assistant LinksUsers of AI assistants with persistent memory features, including Microsoft 365 Copilot, ChatGPT, Claude, Perplexity, and Grok6 steps
- 2026Model Distillation Campaigns Targeting Anthropic ClaudeAnthropic Claude10 steps
- 2025-09-17Poisoned Postmark MCP Server Email ExfiltrationPostmark MCP Server9 steps
- 2025-09GTG-1002 Claude Code Espionage Campaign30 entities in the technology, financial, chemical, and government sectors21 steps
- 2025-07-13Code to Deploy Destructive AI Agent Discovered in Amazon Q VS Code ExtensionAmazon Q VS Code Extension7 steps
- 2025-07SesameOp: Novel backdoor uses OpenAI Assistants API for command and controlOpenAI Assistants API1 step
- 2025-06-25Malware Prototype with Embedded Prompt InjectionLLM malware detectors, LLM malware analysis and reverse engineering tools8 steps
- 2025-06LAMEHUG: Malware Leveraging Dynamic AI-Generated CommandsUkraine's security and defense sector8 steps
- 2025Malicious Models on Hugging FaceHugging Face users6 steps
- 2025Model Distillation Campaigns Targeting Google AI ServicesGoogle AI Services, including Gemini10 steps
- 2024-10-09ProKYC: Deepfake Tool for Account Fraud AttacksKYC verification services8 steps
- 2024-07Storm-2139 Azure OpenAI Guardrail BypassMicrosoft Azure OpenAI Service9 steps
- 2024-05-06LLMjacking: Stolen Cloud Credentials Used to Access Cloud LLM Services.Cloud-Based LLM Services8 steps
- 2024China-Based Artificial Intelligence Companies Conducting Industrial-Scale Distillation Campaigns Against U.S. AI CompaniesModels from Anthropic, OpenAI, Google, and XAI14 steps
- 2023-09-05ShadowRay: Hijacking Exposed Ray ClustersExposed Ray Clusters7 steps
- 2022-12-25Compromised PyTorch Dependency ChainPyTorch3 steps
- 2022-12Attempted Evasion of ML Phishing Webpage Detection SystemCommercial ML Phishing Webpage Detector4 steps
- 2020-10Bypassing ID.me Identity VerificationCalifornia Employment Development Department3 steps
- 2020-04ClearviewAI MisconfigurationClearview AI facial recognition tool4 steps
- 2020Camera Hijack Attack on Facial Recognition SystemShanghai government tax office's facial recognition service8 steps
- 2020VirusTotal PoisoningVirusTotal4 steps
- 2016-03-23Tay PoisoningMicrosoft's Tay AI Chatbot4 steps
Exercises and research · 50
- 2026-06-05Claude Code GitHub Action Secret ExposureClaude Code GitHub Action11 steps
- 2026-05-07RCE Vulnerability in Semantic Kernel Search PluginSemantic Kernel6 steps
- 2026-02-17AI in the Middle: Web-Based AI Services as C2 RelaysEnterprise machines with Grok and Microsoft Copilot access10 steps
- 2026-02-03OpenClaw Command & Control via Prompt InjectionOpenClaw17 steps
- 2026-02-01OpenClaw 1-Click Remote Code ExecutionOpenClaw9 steps
- 2026-01-26Supply Chain Compromise via Poisoned ClawdBot SkillClawdBot (now OpenClaw)12 steps
- 2026-01-25Exposed ClawdBot Control Interfaces Leads to Credential Access and ExecutionClawdBot (now OpenClaw)10 steps
- 2025-09-25ZombieAgent: Data Exfiltration Attack on ChatGPTOpenAI ChatGPT11 steps
- 2025-09-03Model Namespace Reuse Supply Chain AttackUsers of Hugging Face-backed model catalogs, code pipelines, and cloud integrations8 steps
- 2025-08-18Cross-Site Scripting via Prompt Manipulation in Lenovo AI ChatbotLenovo AI chatbot, "Lena"11 steps
- 2025-08-06Prompt-Based Attacks Against Gemini via Calendar InvitationsGoogle Gemini15 steps
- 2025-06-24Data Exfiltration via an MCP Server used by CursorCursor9 steps
- 2025-06-19Living Off AI: Prompt Injection via Jira Service ManagementAtlassian MCP, Jira Service Management8 steps
- 2025-06Data Exfiltration via Agent Tools in Copilot StudioCopilot Studio Customer Service Agent12 steps
- 2025-06Poisoned GGUF Templates: Inference-Time Supply Chain AttackModel registries distributing models in GGUF format13 steps
- 2025-05-24AI ClickFix: Hijacking Computer-Use Agents Using ClickFixClaude Computer-Use Agent8 steps
- 2025-05EchoLeak: Zero-Click Prompt Injection Targeting M365 Copilot for Data ExfiltrationMicrosoft 365 Copilot12 steps
- 2025-04-01Data Exfiltration via Remote Poisoned MCP ToolModel Context Protocol11 steps
- 2025-03-18Rules File Backdoor: Supply Chain Attack on AI Coding AssistantsCursor, GitHub Copilot10 steps
- 2025-03Google Photos AI Model ExtractionGoogle Photos Android App7 steps
- 2025-02-27LLMSmith: RCE Vulnerabilities in LLM-Integrated ApplicationsLLM Integration Frameworks12 steps
- 2025AIKatz: Attacking LLM Desktop ApplicationsLLM Desktop Applications (Claude, ChatGPT, Copilot)12 steps
- 2024-10-24Data Destruction via Indirect Prompt Injection Targeting Claude Computer-UseClaude Computer Use Agent7 steps
- 2024-08-20Data Exfiltration from Slack AI via Indirect Prompt InjectionSlack AI8 steps
- 2024-08-08Financial Transaction Hijacking with M365 Copilot as an InsiderMicrosoft 365 Copilot14 steps
- 2024-06ChatGPT Package HallucinationChatGPT users6 steps
- 2024-03-05Morris II Worm: RAG-Based AttackRAG-based e-mail assistant12 steps
- 2024-02Hacking ChatGPT's Memories with Prompt InjectionOpenAI ChatGPT7 steps
- 2024-02Planting Instructions for Delayed Automatic AI Agent Tool InvocationGoogle Gemini6 steps
- 2024Live Deepfake Image Injection to Evade Mobile KYC VerificationMobile facial authentication service10 steps
- 2023-11-23Google Bard Conversation ExfiltrationGoogle Bard7 steps
- 2023-09-26AI Model Tampering via Supply Chain AttackPrivate Container Registries9 steps
- 2023-08-23Organization Confusion on Hugging FaceHugging Face users16 steps
- 2023-07PoisonGPTHuggingFace Users7 steps
- 2023-05ChatGPT Conversation ExfiltrationOpenAI ChatGPT7 steps
- 2023-01-28Achieving Code Execution in MathGPT via Prompt InjectionMathGPT (https://mathgpt.streamlit.app/)9 steps
- 2023Indirect Prompt Injection Threats: Bing Chat Data PirateMicrosoft Bing Chat5 steps
- 2022-08Web-Scale Data Poisoning: Split-View Attack10 web-scale datasets6 steps
- 2022-07Arbitrary Code Execution with Google ColabGoogle Colab8 steps
- 2021-06-23Confusing Antimalware Neural NetworksKaspersky's Antimalware ML Model9 steps
- 2021-01-18Backdoor Attack on Deep Learning Models in Mobile AppsML-based Android Apps10 steps
- 2020-04-30Attack on Machine Translation ServicesGoogle Translate, Bing Translator, Systran Translate9 steps
- 2020-02Microsoft Edge AI EvasionNew Microsoft AI Product5 steps
- 2020-01-01Face Identification System Evasion via Physical CountermeasuresCommercial Face Identification Service10 steps
- 2020Botnet Domain Generation Algorithm (DGA) Detection EvasionPalo Alto Networks ML-based DGA detection module6 steps
- 2020Evasion of Deep Learning Detector for Malware C&C TrafficPalo Alto Networks malware detection system6 steps
- 2020Microsoft Azure Service DisruptionInternal Microsoft Azure Service8 steps
- 2019-09-09ProofPoint EvasionProofPoint Email Protection System5 steps
- 2019-09-07Bypassing Cylance's AI Malware DetectionCylancePROTECT, Cylance Smart Antivirus6 steps
- 2019-08-22GPT-2 Model ReplicationOpenAI GPT-25 steps
Reported in the news · AI Incident Database · 132
- 2026-10-30Attacker Reportedly Used AI Agents to Breach Blockchain Analytics Firm Chainalysis and Access Customer DataNot yet classified
- 2026-10-29Defense Contractor's Personal AI Coding Agent Reportedly Sent U.S. Defense Department Credentials to a Third-party WebsiteNot yet classified
- 2026-10-26Cybersecurity Firm Cathedral Reportedly Used AI Agents to Access Anthropic's Internal Business SystemsNot yet classified
- 2026-10-20AI Agents Operated by Safety-Testing Firm Trajectory Labs Attempted Website Exploits and Impersonated Staff at AI Research Organization Epoch AINot yet classified
- 2026-10-17China-Linked Espionage Group VAPOR PANDA Reportedly Used ARTEX AI Agents to Breach Government and Corporate Systems WorldwideNot yet classified
- 2026-10-15ARTEX AI Agents Reportedly Used Stolen Cloud Backup Credentials to Access ConnectWise Customers' Microsoft 365 EnvironmentsNot yet classified
- 2026-10-15AI Agents Under Evaluation Reportedly Accessed Personal and Confidential Records at Indian OrganizationsNot yet classified
- 2026-10-04Tencent-Linked AI Agents Reportedly Used Exposed Credentials to Retrieve Data From Alibaba's Mapping ServiceNot yet classified
- 2026-09-27Attackers Reportedly Used ARTEX, an AI Security-Testing Tool, to Steal Data from South Korean Financial InstitutionsNot yet classified
- 2026-09-25Cybercriminal Reportedly Used Cairn AI Agents to Breach More Than 100 Company WebsitesNot yet classified
- 2026-09-25AI Agent Appearing to Originate from OpenAI Reportedly Attempted to Hack U.S. Department of Education Civil Rights WebsiteNot yet classified
- 2026-09-14AI Agent Reportedly Exploited Application Vulnerabilities to Modify Personal Data and Access Invoices in Breach Reported to Spanish Data Protection AuthorityNot yet classified
- 2026-09-09Early Claude Opus 4.6 Checkpoint Reportedly Gained Unauthorized Admin Access to Third-Party System During Cybersecurity EvaluationNot yet classified
- 2026-07-30Anthropic Research Model Reportedly Scanned 9,000 Targets and Compromised Real Company's Application During EvaluationNot yet classified
- 2026-07-30Claude Mythos 5 Reportedly Published Malicious PyPI Package That Compromised Real Security Company During EvaluationNot yet classified
- 2026-07-30Claude Opus 4.7 Reportedly Compromised Real Company's Production Infrastructure During Cybersecurity EvaluationNot yet classified
- 2026-07-11OpenAI Models Reportedly Compromised Hugging Face Production Infrastructure During Cybersecurity EvaluationNot yet classified
- 2026-07-09Threat Actor Reportedly Used Hermes AI Agent for Unattended Post-Compromise Activity in Thailand's Ministry of Finance NetworkNot yet classified
- 2026-07-08Threat Actor Reportedly Used AI-Assisted Workflows to Compromise AWS Environment for ExtortionNot yet classified
- 2026-07-01Suspected China-Linked Hackers Reportedly Used AI Agents to Compromise Taiwanese Government Systems and Exfiltrate Personnel RecordsNot yet classified
- 2026-07-01LLM-Driven Ransomware Operator Dubbed JADEPUFFER Reportedly Targeted Production DatabaseNot yet classified
- 2026-06-17OpenAI Training Agent Reportedly Retrieved Credentials and Restricted Data from a Government Statistics WebsiteNot yet classified
- 2026-05-31Meta AI Support Bot Reportedly Enabled Takeovers of High-Profile Instagram AccountsNot yet classified
- 2026-05-30COEMPT Quality Assurance Engineers Allegedly Violated Indian CBSE Student Data Privacy Rights by Processing It with Google GeminiAI system security vulnerabilities and attacks
- 2026-05-28AI Agents Reportedly Attempted to Hack Canada's Federal Archives Website While Retrieving Historical Divorce RecordsNot yet classified
- 2026-05-24OpenAI-Linked AI Agents Reportedly Used German Programming Wiki DSEWiki for Coordination and Restriction EvasionNot yet classified
- 2026-05-20Google Gemini 3.5 Coding Agent Reportedly Caused Production Portal Outage and Generated Fabricated Recovery RecordsNot yet classified
- 2026-05-19Attacker Reportedly Used AI Agents to Breach Healthcare Organizations and Steal Medical Records in ArgentinaNot yet classified
- 2026-05-13Hidden Prompt Injection in Brazilian Labor-Court Petition Reportedly Tried to Manipulate GalileuAI system security vulnerabilities and attacks
- 2026-05-10Threat Actor Reportedly Used LLM Agent to Exfiltrate Internal Database After Compromising marimo Python NotebookNot yet classified
- 2026-05-01Google Gemini Reportedly Accessed Three Real Companies' Protected Systems During Cybersecurity EvaluationNot yet classified
- 2026-05-01Google Alleged Outsider Enterprise Used Gemini to Create Phishing Sites for Text-Message ScamsNot yet classified
- 2026-04-30AI Agent Powered by Claude Opus 4.6 Reportedly Exploited Gym Booking API and Removed Another Member From WaitlistNot yet classified
- 2026-04-28Ahmedabad Aadhaar Fraud Racket Reportedly Used Purportedly AI-Generated Deepfakes to Change Businessman's Linked Mobile NumberFraud, scams, and targeted manipulation
- 2026-04-08Aurora Ransomware Operator Reportedly Used Cursor Agent to Assist Intrusions Against Multiple OrganizationsNot yet classified
- 2026-04-07Claude Mythos Preview Reportedly Posted Sandbox Exploit Details to Public Websites During Anthropic EvaluationNot yet classified
- 2026-03-27OpenAI Research Agent Reportedly Compromised an Internal Chip-Design Server During an EvaluationNot yet classified
- 2026-03-18Meta Internal AI Agent Reportedly Gave Advice That Allegedly Exposed Sensitive Data to Unauthorized EmployeesAI system security vulnerabilities and attacks
- 2026-02-23Anthropic Said DeepSeek, Moonshot, and MiniMax Used Fraudulent Accounts and Proxies to Illicitly Distill Claude Capabilities at ScaleAI system security vulnerabilities and attacks
- 2026-02-19Grok Reportedly Disclosed Adult Performer Siri Dahl's Legal Name and Birthdate, Allegedly Contributing to Doxxing and HarassmentCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2026-02-17Unknown Actor Reportedly Exploited Cline's Claude-Powered GitHub Issue-Triage Workflow and Published an Unauthorized Cline CLI PackageNot yet classified
- 2026-01-31Moltbook Database Exposure Allegedly Revealed Users' Private Communications and API Authentication TokensAI system security vulnerabilities and attacks
- 2025-12-01Anthropic's Claude Was Reportedly Jailbroken To Allegedly Help Steal Sensitive Mexican Government DataCyberattacks, weapon development or use, and mass harm
- 2025-11-19Secret Desires AI Platform Reportedly Exposed Nearly Two Million Sensitive Images in Cloud Storage LeakCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2025-10-10OpenAI ChatGPT Models Reportedly Jailbroken to Provide Chemical, Biological, and Nuclear Weapons InstructionsCyberattacks, weapon development or use, and mass harm
- 2025-09-11Purported AI-Generated Deepfake of Irish Fine Gael Presidential Candidate Heather Humphreys Used in Fake Investment Videos on Meta PlatformsFraud, scams, and targeted manipulation
- 2025-08-27Anthropic Reportedly Identifies AI Misuse in Extortion Campaigns, North Korean IT Schemes, and Ransomware SalesCyberattacks, weapon development or use, and mass harm
- 2025-08-21Malicious Nx npm Packages Reportedly Weaponize AI Coding Agents for Data ExfiltrationAI system security vulnerabilities and attacks
- 2025-07-31Reported Public Exposure of Over 100,000 LLM Conversations via Share Links Indexed by Search Engines and ArchivedCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2025-07-17North Korea's Kimsuky Group Reportedly Uses AI-Generated Military ID Deepfakes in Phishing CampaignFraud, scams, and targeted manipulation
- 2025-07-15CISA Acting Director Reportedly Uploaded Sensitive Government Documents to Public ChatGPT InstanceCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2025-07-10LAMEHUG Malware Reportedly Integrates Large Language Model for Real-Time Command Generation in a Purported APT28-Linked CyberattackCyberattacks, weapon development or use, and mass harm
- 2025-07-09Urban VPN Proxy Browser Extension Reportedly Harvested and Sold Private AI Chatbot Conversations via Silent UpdateCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2025-07-04Microsoft 365 Copilot Vulnerability Allegedly Allowed File Access Without Audit Log EntryAI system security vulnerabilities and attacks
- 2025-06-30McDonald's McHire AI Recruitment Platform Reportedly Exposed Data of 64 Million Applicants via Default Login and API VulnerabilityCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2025-06-28Reportedly Sustained Multi-Celebrity Deepfake Persona Scam Targeting Vulnerable Southampton ResidentFraud, scams, and targeted manipulation
- 2025-06-26Reported Student Misuse of ChatGPT, StudyX, and Gemini to Obtain Answers During Vietnam's 2025 National High School Graduation ExamFraud, scams, and targeted manipulation
- 2025-06-22North Korea-Linked Actors Allegedly Use AI Executive Deepfakes in Zoom Phishing Targeting Web3 EmployeeFraud, scams, and targeted manipulation
- 2025-06-19Reported AI-Generated Video Call Impersonation of Cryptocurrency Analyst Leads to Alleged Malware Installation and Account TheftFraud, scams, and targeted manipulation
- 2025-06-18Purported Deepfake of Sri Lankan President Anura Kumara Dissanayake Promotes Alleged Fraudulent Government Investment SchemeFraud, scams, and targeted manipulation
- 2025-05-17AI Chatbot Allegedly Used to Research Explosive Materials in Palm Springs Fertility Clinic BombingCyberattacks, weapon development or use, and mass harm
- 2025-05-15FBI Reports Ongoing Vishing and Smishing Campaign Allegedly Targeting Government Officials Using Purportedly AI-Generated VoicesFraud, scams, and targeted manipulation
- 2025-05-09Serviceaide AI Platform Implicated in Health Data Exposure Affecting 483,000 Catholic Health PatientsCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2025-04-29Meta AI App Reportedly Publishes Personal Chats Without Users Fully RealizingCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2025-04-17ChatGPT Was Alleged to Have Aided Planning of Florida State University Mass ShootingCyberattacks, weapon development or use, and mass harm
- 2025-04-16Microsoft Reportedly Blocks 1.6 Million Bot Signup Attempts Per Hour Amid Global AI-Driven Fraud SurgeFraud, scams, and targeted manipulation
- 2025-04-11Reportedly Unsafe Deployment of Llama.cpp Reveals Interactive AI-Generated CSAM Roleplay PromptsCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2025-04-09Jailbroken Lovable AI Allegedly Used to Generate and Host Phishing Pages, Steal Credentials, and Bypass SecurityFraud, scams, and targeted manipulation
- 2025-04-08Deepfake Job Applicant Allegedly Used AI Tools to Apply for Remote Role at U.S. Security StartupFraud, scams, and targeted manipulation
- 2025-03-31GenNomis AI Database Reportedly Exposes Nearly 100,000 Deepfake and Nudify Images in Public BreachCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2025-03-18Alleged Fraudulent Prompts via AIXBT Dashboard Led Purported AI Trading Agent to Transfer 55.5 ETH from Simulacrum WalletAI system security vulnerabilities and attacks
- 2025-03-12Alleged ChatGPT Misuse by Contractor Leads to Reported Data Exposure in New South Wales Resilient Homes ProgramCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2025-03-12Attackers Reportedly Deployed Simulated AI Support Chatbot to Trick Instagram Business Users into Adding Malicious 2FA LoginFraud, scams, and targeted manipulation
- 2025-03-05ChatGPT Was Allegedly Consulted About Attack Options Before Teen Attempted to Stab Officer at Tira, Israel, Police StationCyberattacks, weapon development or use, and mass harm
- 2025-03-05Google Reports Alleged Gemini-Generated Terrorism and Child Exploitation to Australian eSafety CommissionFraud, scams, and targeted manipulation
- 2025-03-04Reported Deepfake Influencers on TikTok Allegedly Used to Promote Fraudulent Wellness ProductsFraud, scams, and targeted manipulation
- 2025-03-04Phishers Allegedly Using AI-Generated Video of YouTube CEO Neal Mohan to Target CreatorsFraud, scams, and targeted manipulation
- 2025-03-02Deepfake Video of Indonesian President Prabowo Subianto and Other Officials Reportedly Used in Scam to Defraud Citizens Across 20 ProvincesFraud, scams, and targeted manipulation
- 2025-02-28Alleged Inclusion of 12,000 Live API Keys in LLM Training Data Reportedly Poses Security RisksCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2025-02-23Alleged FraudGPT-Enabled Phishing Attack Spoofs ChatGPT Subscription Service to Steal CredentialsFraud, scams, and targeted manipulation
- 2025-02-20Cybercriminals Reportedly Exploited Google’s G.Co Subdomain and Spoofed Caller ID in AI-Driven Phishing Attack on Hack Club FounderFraud, scams, and targeted manipulation
- 2025-02-18Alleged Use of Purported AI-Generated Identities to Defraud FTX Claims Buyers of $5.6MFraud, scams, and targeted manipulation
- 2025-02-17AI-Driven Phishing Scam Uses Deepfake Robocalls to Target Gmail Users in Credential Theft CampaignFraud, scams, and targeted manipulation
- 2025-02-10AI-Generated Voice Purporting to Be Daughter Allegedly Used to Coerce $2,000 from Colorado MotherFraud, scams, and targeted manipulation
- 2025-01-10AI-Assisted Ransomware Campaign by FunkSec Allegedly Targets Over 80 VictimsFraud, scams, and targeted manipulation
- 2024-12-27ChatGPT Reportedly Referenced During Las Vegas Cybertruck Explosion PlanningCyberattacks, weapon development or use, and mass harm
- 2024-12-26Meta AI Bug in Deployed Service Reportedly Allowed Potential Access to Other Users' Prompts and ResponsesCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2024-12-26Alleged Deepfake of Luis Suárez Used to Scam Uruguayans in Fake Investment SchemeFraud, scams, and targeted manipulation
- 2024-12-19Global Cybercrime Network Storm-2139 Allegedly Exploits AI to Generate Deepfake ContentFraud, scams, and targeted manipulation
- 2024-12-19Romance Scammer 'Alla Morgan' Allegedly Exploits Deepfake Technology to Defraud Victim of £17,000Fraud, scams, and targeted manipulation
- 2024-12-16HTML/Nomani Deepfake Phishing Campaigns Allegedly Use AI-Generated Content to Defraud Social Media UsersFraud, scams, and targeted manipulation
- 2024-12-16AI Voice Cloning of Ari Melber Allegedly Exploited in Scam Targeting Elderly WomanFraud, scams, and targeted manipulation
- 2024-11-25ISWAP Reportedly Used AI Guidance to Breach Nigerian Military Base Defenses by MotorcycleNot yet classified
- 2024-11-24Deepfake Videos Allegedly Used to Defraud Canadian Immigrants Out of Thousands of DollarsFraud, scams, and targeted manipulation
- 2024-11-01Deepfake Videos of Barbara O’Neill Allegedly Used in Health Scam Targeting Social Media UsersFraud, scams, and targeted manipulation
- 2024-10-09ProKYC Tool Allegedly Facilitates Deepfake-Based Account Fraud on Cryptocurrency ExchangesFraud, scams, and targeted manipulation
- 2024-10-08Alleged AI-Powered Call Center Breach Exposes Over 10 Million Conversations in the Middle EastCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2024-10-02Fake AI 'Nudify' Sites Reportedly Linked to Malware Distribution by Russian Hacker Collective FIN7Fraud, scams, and targeted manipulation
- 2024-09-28Jennifer Aniston’s Likeness Allegedly Exploited in Purported Deepfake Collagen Supplement PromotionFraud, scams, and targeted manipulation
- 2024-09-25ChatGPT Reportedly Introduces Errors in Critical Child Protection Court ReportCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2024-09-13Criminal Group Uses AI Deepfake Technology to Steal Personal Data in Hangzhou, ZhejiangFraud, scams, and targeted manipulation
- 2024-09-13Purported Deepfake of Brian May Allegedly Exploited in Scam Offering Fake Queen Backstage TicketsFraud, scams, and targeted manipulation
- 2024-09-09AI 'Nudify' Apps Used as Tools for Blackmail and ExtortionFraud, scams, and targeted manipulation
- 2024-08-23Generative AI Allegedly Used to Facilitate $255,000 Real Estate Fraud SchemeFraud, scams, and targeted manipulation
- 2024-07-04China Targets AI-Driven Fraud and Deepfake Scandals with New CrackdownsFraud, scams, and targeted manipulation
- 2024-07-01Scammers Allegedly Use Deepfake Technology to Pose as Leonor, Princess of Asturias, in Fraud SchemeFraud, scams, and targeted manipulation
- 2024-07-01OpenAI's ChatGPT Mac App Stored User Data in Unencrypted Text FilesAI system security vulnerabilities and attacks
- 2024-05-06Alleged LLMjacking Targets AI Cloud Services with Stolen CredentialsFraud, scams, and targeted manipulation
- 2024-04-10Reported Audio Deepfake Impersonating CEO Karim Toubba Targets LastPass Employee via WhatsAppFraud, scams, and targeted manipulation
- 2024-04-08Deepfake Porn Sites Use Breeze Liu's Image Without ConsentFraud, scams, and targeted manipulation
- 2024-03-26Leonardo AI's Platform Alleged to Have Been Used for Creating Nonconsensual Celebrity DeepfakesFraud, scams, and targeted manipulation
- 2024-03-01Deepfake ID Sales Allegedly Persist as OnlyFake Reportedly Relaunches with New Fraud ToolsFraud, scams, and targeted manipulation
- 2024-02-14AI Romance Apps Reportedly Compromise User Privacy for Data HarvestingCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2024-01-30Alleged ChatGPT Account Compromise Reportedly Led to Unintended Data ExposureCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2023-12-01Underground Market for LLMs Powers Malware and Phishing ScamsCyberattacks, weapon development or use, and mass harm
- 2023-10-20Reported Use of AI Voice and Identity Manipulation in the Ongoing 'Phantom Hacker' Fraud SchemeFraud, scams, and targeted manipulation
- 2023-10-15Aledo High School Student Allegedly Generates and Distributes Deepfake Nudes of Seven Female ClassmatesFraud, scams, and targeted manipulation
- 2023-08-27Deepfake Voice Exploit Compromises Retool's Cloud ServicesFraud, scams, and targeted manipulation
- 2023-06-30Grant Reviewers Reportedly Fed Applications into Generative AI to Produce Reports, Allegedly Breaching ConfidentialityFraud, scams, and targeted manipulation
- 2023-06-22Bing Chat Solved CAPTCHAs with Image Analysis Feature Despite SafeguardsAI system security vulnerabilities and attacks
- 2023-04-14Quebec Man Sentenced for Having Used Deepfake Technology to Create Synthetic Child PornographyFraud, scams, and targeted manipulation
- 2023-03-31ChatGPT Reportedly Banned by Italian Authority Due to OpenAI's Purported Lack of Legal Basis for Data Collection and Age VerificationCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2023-03-20ChatGPT Reportedly Exposed Users' Private Data Reportedly Due to BugCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2023-03-11ChatGPT Reportedly Implicated in Samsung Data Leak of Source Code and Meeting NotesCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2023-02-08Bing Chat's Initial Prompts Revealed by Early Testers Through Prompt InjectionAI system security vulnerabilities and attacks
- 2022-12-21ChatGPT Abused to Develop Malicious SoftwaresCyberattacks, weapon development or use, and mass harm
- 2022-09-15GPT-3-Based Twitter Bot Hijacked Using Prompt Injection AttacksAI system security vulnerabilities and attacks
- 2021-10-15McDonald's AI Drive-Thru Allegedly Collected Biometric Customer Data without Consent, Violating BIPACompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2020-06-08Kristen Bell Deepfaked in Non-Consensual AI-Generated PornographyFraud, scams, and targeted manipulation
- 2020-01-01Quantum AI and Related AI-Themed Investment Scams Reportedly Used Deepfake Endorsements and Spoofed Media Sites to Solicit InvestmentsFraud, scams, and targeted manipulation
- 2019-02-14GPT-2 Reportedly Reproduced Personal Data from Its Training DataCompromise of privacy by obtaining, leaking or correctly inferring sensitive information
- 2016-12-22Reinforcement Learning Reward Functions in Video GamesAI pursuing its own goals in conflict with human goals or values
Titles from the AI Incident Database (incidentdatabase.ai), snapshot of 2026-11-16, under CC BY-SA 4.0. Incidents that involve generative AI and a security harm only.
From MITRE ATLAS 2026.10 (Apache-2.0). ShieldMap lists the facts and the technique chain of each case study and links to ATLAS for the full write-up; it does not copy the steps.