All Records

Reverse Shell - AI Security Technique

Reverse Shell is AI attack method AML.T0072 with evidence level: realized. It includes 3 case studies.

Adversarial AI Attacks - AI Security Technique

Adversarial AI Attacks is AI attack method AML.T0017.000 with evidence level: demonstrated. It includes 4 case studies.

Software Tools - AI Security Technique

Software Tools is AI attack method AML.T0016.001 with evidence level: realized. It includes 3 case studies.

Adversarial AI Attack Implementations - AI Security Technique

Adversarial AI Attack Implementations is AI attack method AML.T0016.000 with evidence level: realized. It includes 3 case studies.

Poisoned Postmark MCP Server Email Exfiltration - AI Case Study

Poisoned Postmark MCP Server Email Exfiltration - AI Case Study is an AI security case study covering type: incident; target: Postmark MCP Server; actor: Unk...

ChatGPT Package Hallucination - AI Case Study

ChatGPT Package Hallucination - AI Case Study is an AI security case study covering type: exercise; target: ChatGPT users; actor: Vulcan Cyber, Lasso Securit...

Evasion of Deep Learning Detector for Malware C&C Traffic - AI Case Study

Evasion of Deep Learning Detector for Malware C&C Traffic - AI Case Study is an AI security case study covering type: exercise; target: Palo Alto Networks ma...

Black-Box Optimization - AI Security Technique

Black-Box Optimization is AI attack method AML.T0043.001 with evidence level: demonstrated. It includes 7 mitigations; 2 case studies.

Data Destruction via Indirect Prompt Injection Targeting Claude Computer-Use - AI Case Study

Data Destruction via Indirect Prompt Injection Targeting Claude Computer-Use - AI Case Study is an AI security case study covering type: exercise; target: Cl...

Live Deepfake Image Injection to Evade Mobile KYC Verification - AI Case Study

Live Deepfake Image Injection to Evade Mobile KYC Verification - AI Case Study is an AI security case study covering type: exercise; target: Mobile facial au...

ChatGPT Conversation Exfiltration - AI Case Study

ChatGPT Conversation Exfiltration - AI Case Study is an AI security case study covering type: exercise; target: OpenAI ChatGPT; actor: Embrace The Red. It in...

ProofPoint Evasion - AI Case Study

ProofPoint Evasion - AI Case Study is an AI security case study covering type: exercise; target: ProofPoint Email Protection System; actor: Researchers at Si...

Risks from data (Risks of illegal collection and use of data)

Risks from data (Risks of illegal collection and use of data) is an AI risk in 2. Privacy & Security focused on 2.1 > Compromise of privacy by leaking or cor...

Black-Box Transfer - AI Security Technique

Black-Box Transfer is AI attack method AML.T0043.002 with evidence level: demonstrated. It includes 4 mitigations; 3 case studies.

1.2 > Exposure to toxic content AI Risks

1.2 > Exposure to toxic content groups 63 AI risk records under 1. Discrimination & Toxicity.

AI's persuasive capabilities are misused to gain influence and promote harmful ideologies

AI's persuasive capabilities are misused to gain influence and promote harmful ideologies is an AI risk in 4. Malicious Actors & Misuse focused on 4.1 > Disi...

Acquire Public AI Artifacts - AI Security Technique

Acquire Public AI Artifacts is AI attack method AML.T0002 with evidence level: realized. It includes 1 mitigation; 3 case studies.

Bypassing Cylance's AI Malware Detection - AI Case Study

Bypassing Cylance's AI Malware Detection - AI Case Study is an AI security case study covering type: exercise; target: CylancePROTECT, Cylance Smart Antiviru...

Generative AI User Data

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

Craft Adversarial Data - AI Security Technique

Craft Adversarial Data is AI attack method AML.T0043 with evidence level: realized. It includes 8 mitigations; 1 case study.