Please enable JavaScript.
Coggle requires JavaScript to display documents.
AI Security - Coggle Diagram
AI Security
Ethical Considerations
Algorithmic bias
Explainability and transparency
Human oversight and control
Potential job displacement
Informed consent
Safety and reliability
Dual use of AI technology
Long-term societal impact
Regulatory compliance
Industries Affected
Healthcare
Finance
Automotive
Retail
Manufacturing
Energy and Utilities
Telecommunications
Education
Agriculture
Authentication
Knowledge-Based
Possession-Based
Inherence-Based
Physical
Behavioral
Location/Contextual
Multi-Factor & Adaptive Authentication
Access Control
Role-Based Access Control (RBAC)
Attribute-Based Access Control (ABAC)
Mandatory Access Control (MAC)
Discretionary Access Control (DAC)
Context-Based Access Control (CBAC)
Model Security
Threats
Model Inversion
Model Stealing
Model Extraction
Prompt Injection
IP Theft
Parameter Tampering
Output Manipulation
Reverse Engineering Inputs
Defense Techniques
Data Sanitization
Algorithm Robustness Enhancement
Security Assessment Mechanism
Privacy Preserving Techniques
Model Hardening
Regularization
Adversarial Training
Detection Techniques
Intrusion Detection Systems (IDS)
Anomaly Detection
Watermarking
Logging, Monitoring, and Alerting (LMA) framework - Cloud Observability Service (COS)
Prevention Approach
Re-training the model
Employing differential privacy
Input and output perturbations
Model modification
Data Security
Threats
Data Breaches
Adversarial Attacks
Model Inversion
Privacy Violation
Data Poisoning
Strategies for Securing Data
Encryption
Access Control
Anonymization and Differential Privacy
Model Robustness
Continuous Monitoring
Ethical AI Practices