Negative impact “Adverse impact on human perception and judgment caused by disinformation”

This page provides the security targets of negative impacts “Adverse impact on human perception and judgment caused by disinformation” in the external influence aspect in the AI Security Map, as well as the attacks and factors that cause them, and the corresponding defense methods and countermeasures.

Security target

• Non-consumer

Attack or cause

• Integrity violation
• Availability breach
• Explainability violation
• Degradation of output fairness
• Reliability violation
• Deepfake
• Social engineering attack

Defensive method or countermeasure

• AI alignment
• Watermarking for generative AI
• Identification of AI-generated output
• Detection of disinformation
• Deepfake detection
• Education and follow-up

References