Record summary
A quick snapshot of what this page covers.
Techniques18Attack methods connected to this risk.
Mitigations15Defenses that may help with related attacks.
Domain2. Privacy & SecurityThe broad risk area this belongs to.
Risk profile
How this risk is described and categorized.
Domain2. Privacy & Security
Subdomain2.2 > AI system security vulnerabilities and attacks
Entity1 - Human
Intent1 - Intentional
Timing2 - Post-deployment
CategoryModel Attacks
SubcategoryInference Attacks
Suggested mitigations
Defenses that may help with related attacks.
Restrict Number of AI Model Queries
Business and Data UnderstandingDeployment+1 more
Control Access to AI Models and Data in Production
DeploymentMonitoring and Maintenance
AI Telemetry Logging
DeploymentMonitoring and Maintenance
Control Access to AI Models and Data at Rest
Business and Data UnderstandingData Preparation+2 more
Encrypt Sensitive Information
Data PreparationML Model Engineering+1 more
Sanitize Training Data
Business and Data UnderstandingData Preparation+1 more
Verify AI Artifacts
Business and Data UnderstandingData Preparation+1 more
Maintain AI Dataset Provenance
Data PreparationBusiness and Data Understanding
Memory Hardening
ML Model EngineeringDeployment+1 more
Model Hardening
Data PreparationML Model Engineering
Use Ensemble Methods
ML Model Engineering
Input Restoration
Data PreparationML Model Evaluation+2 more
Adversarial Input Detection
Data PreparationML Model Engineering+3 more
Validate AI Model
ML Model EvaluationMonitoring and Maintenance
Passive AI Output Obfuscation
DeploymentML Model Evaluation
Source
Research source for this risk, when available.
Included resource
Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems
Original source
MIT AI Risk Repository
Open the public repository used for AI risk records and taxonomy fields.