Record summary
A quick snapshot of what this page covers.
Techniques3Attack methods connected to this risk.
Mitigations1Defenses that may help with related attacks.
Domain2. Privacy & SecurityThe broad risk area this belongs to.
Risk profile
How this risk is described and categorized.
"Adversarial attacks can affect not only the model’s output but also its corresponding explanation. Current adversarial optimization techniques can intro- duce imperceptible noise to the input image, so that the model’s output does not change but the corresponding explanation is arbitrarily manipulated [61]. Such manipulations are harder to notice, as they are less commonly known compared to standard adversarial attacks targeting the model’s output."
Domain2. Privacy & Security
Subdomain2.2 > AI system security vulnerabilities and attacks
Entity1 - Human
Intent1 - Intentional
Timing3 - Other
CategoryModel Evaluations (Interpretability/Explainability)
SubcategoryAdversarial attacks targeting explainable AI techniques
Suggested mitigations
Defenses that may help with related attacks.
Limit Public Release of Information
Business and Data Understanding
Source
Research source for this risk, when available.
Included resource
Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems
Original source
MIT AI Risk Repository
Open the public repository used for AI risk records and taxonomy fields.