All Records

Societal Harm - AI Security Technique

Societal Harm is AI attack method AML.T0048.002 with evidence level: realized. It includes 1 case study; 2 related AI risks.

Acquire Infrastructure - AI Security Technique

Acquire Infrastructure is AI attack method AML.T0008 with evidence level: realized. It includes 2 case studies.

Active Scanning - AI Security Technique

Active Scanning is AI attack method AML.T0006 with evidence level: realized. It includes 2 case studies.

6.2 > Increased inequality and decline in employment quality AI Risks

6.2 > Increased inequality and decline in employment quality groups 51 AI risk records under 6. Socioeconomic and Environmental.

Compromising privacy or security by correctly inferring sensitive information

Compromising privacy or security by correctly inferring sensitive information is an AI risk in 2. Privacy & Security focused on 2.1 > Compromise of privacy b...

7.6 > Multi-agent risks AI Risks

7.6 > Multi-agent risks groups 52 AI risk records under 7. AI System Safety, Failures, & Limitations.

Discover AI Model Outputs - AI Security Technique

Discover AI Model Outputs is AI attack method AML.T0063 with evidence level: demonstrated. It includes 4 mitigations; 2 case studies.

Tay Poisoning - AI Case Study

Tay Poisoning - AI Case Study is an AI security case study covering type: incident; target: Microsoft's Tay AI Chatbot; actor: 4chan Users. It includes 4 cas...

5.1 > Overreliance and unsafe use AI Risks

5.1 > Overreliance and unsafe use groups 53 AI risk records under 5. Human-Computer Interaction.

Avenues for exploiting user trust and accessing more private information

Avenues for exploiting user trust and accessing more private information is an AI risk in 5. Human-Computer Interaction focused on 5.1 > Overreliance and uns...

Privacy

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

Data usage rights restrictions

Data usage rights restrictions is an AI risk focused on X.1 > Excluded. It is most relevant during 4 - Not coded.

Search Application Repositories - AI Security Technique

Search Application Repositories is AI attack method AML.T0004 with evidence level: demonstrated. It includes 1 mitigation; 3 case studies.

LAMEHUG: Malware Leveraging Dynamic AI-Generated Commands - AI Case Study

LAMEHUG: Malware Leveraging Dynamic AI-Generated Commands - AI Case Study is an AI security case study covering type: incident; target: Ukraine’s security an...

LLM Jacking - AI Case Study

LLM Jacking - AI Case Study is an AI security case study covering type: incident; target: Cloud-Based LLM Services; actor: Unknown. It includes 7 case steps.

Microsoft Edge AI Evasion - AI Case Study

Microsoft Edge AI Evasion - AI Case Study is an AI security case study covering type: exercise; target: New Microsoft AI Product; actor: Azure Red Team. It i...

Camera Hijack Attack on Facial Recognition System - AI Case Study

Camera Hijack Attack on Facial Recognition System - AI Case Study is an AI security case study covering type: incident; target: Shanghai government tax offic...

Opaque Data Collection

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

Pre-processing Tools

Pre-processing Tools is an AI risk in 2. Privacy & Security focused on 2.2 > AI system security vulnerabilities and attacks. It is most relevant during 1 - P...

AI Supply Chain Compromise - AI Security Technique

AI Supply Chain Compromise is AI attack method AML.T0010 with evidence level: realized. It includes 3 mitigations; 1 case study; 1 related AI risk.