All Records

Virtualization/Sandbox Evasion - AI Security Technique

Virtualization/Sandbox Evasion is AI attack method AML.T0097 with evidence level: realized. It includes 1 case study; 2 related AI risks.

Tool Definitions - AI Security Technique

Tool Definitions is AI attack method AML.T0084.001 with evidence level: demonstrated. It includes 1 case study; 5 related AI risks.

Control Access to AI Models and Data in Production - AI Mitigation

Control Access to AI Models and Data in Production is an AI mitigation for Deployment, Monitoring and Maintenance, grouped as Policy. It is connected to 11 a...

Use Ensemble Methods - AI Mitigation

Use Ensemble Methods is an AI mitigation for ML Model Engineering, grouped as Technical - ML. It is connected to 11 attack methods.

Passive AI Output Obfuscation - AI Mitigation

Passive AI Output Obfuscation is an AI mitigation for Deployment, ML Model Evaluation, grouped as Technical - ML. It is connected to 11 attack methods.

6.5 > Governance failure AI Risks

6.5 > Governance failure groups 56 AI risk records under 6. Socioeconomic and Environmental.

Proprietary data

Proprietary data is an AI risk in 2. Privacy & Security focused on 2.1 > Compromise of privacy by leaking or correctly inferring sensitive information. It is...

CVE-2021-3129 - Laravel Ignition

CVE-2021-3129 is a vulnerability profile for Laravel Ignition listed by CISA as known exploited. It includes 1 related AI attack method; 1 related case study.

Propaganda - Digital impersonations

Propaganda - Digital impersonations is an AI risk in 4. Malicious Actors & Misuse focused on 4.3 > Fraud, scams, and targeted manipulation. It is most releva...

Issues on External Tools

Issues on External Tools is an AI risk in 2. Privacy & Security focused on 2.2 > AI system security vulnerabilities and attacks. It is most relevant during 3...

Rules File Backdoor: Supply Chain Attack on AI Coding Assistants - AI Case Study

Rules File Backdoor: Supply Chain Attack on AI Coding Assistants - AI Case Study is an AI security case study covering type: exercise; target: Cursor, GitHub...

Planting Instructions for Delayed Automatic AI Agent Tool Invocation - AI Case Study

Planting Instructions for Delayed Automatic AI Agent Tool Invocation - AI Case Study is an AI security case study covering type: exercise; target: Google Gem...

ProKYC: Deepfake Tool for Account Fraud Attacks - AI Case Study

ProKYC: Deepfake Tool for Account Fraud Attacks - AI Case Study is an AI security case study covering type: incident; target: KYC verification services; acto...

Web-Scale Data Poisoning: Split-View Attack - AI Case Study

Web-Scale Data Poisoning: Split-View Attack - AI Case Study is an AI security case study covering type: exercise; target: 10 web-scale datasets; actor: Resea...

Botnet Domain Generation Algorithm (DGA) Detection Evasion - AI Case Study

Botnet Domain Generation Algorithm (DGA) Detection Evasion - AI Case Study is an AI security case study covering type: exercise; target: Palo Alto Networks M...

White-Box Optimization - AI Security Technique

White-Box Optimization is AI attack method AML.T0043.000 with evidence level: demonstrated. It includes 6 mitigations; 2 case studies.

VirusTotal Poisoning - AI Case Study

VirusTotal Poisoning - AI Case Study is an AI security case study covering type: incident; target: VirusTotal; actor: Unknown. It includes 4 case steps.

Evasion attack

Evasion attack is an AI risk in 2. Privacy & Security focused on 2.2 > AI system security vulnerabilities and attacks. It is most relevant during 2 - Post-de...

Malware Prototype with Embedded Prompt Injection - AI Case Study

Malware Prototype with Embedded Prompt Injection - AI Case Study is an AI security case study covering type: incident; target: LLM malware detectors, LLM mal...

Gather Victim Identity Information - AI Security Technique

Gather Victim Identity Information is AI attack method AML.T0087 with evidence level: realized. It includes 3 case studies.