CERT-In AI Security Auditing Guidelines
In News: The Indian Computer Emergency Response Team (CERT-In) has issued the 'Artificial Intelligence (AI) Security Auditing Guidelines', a comprehensive framework to regulate, monitor, and audit risks associated with agentic AI systems and their autonomous capabilities.
What is Agentic AI?
- Traditional Generative AI primarily responds to user prompts with text, images, or code (passive response).
- Agentic AI refers to autonomous artificial intelligence systems capable of taking independent actions to achieve designated goals.
- These 'agents' can:
- Interact with other software
- Execute commands
- Browse the internet
- Manipulate files
- Operate with minimal human intervention
> The key distinction: Generative AI responds; Agentic AI acts.
Key Features of the Guidelines
1. Risk-Tiering System
- Classifies AI systems into four risk tiers: Low, Moderate, High, and Critical.
- Classification is based on:
- Impact severity
- Data sensitivity
- Decision-making autonomy
- The tier dictates the depth, frequency, and independence of the audit.
2. Comprehensive Audit Domains
Covers nine core domains across the entire AI lifecycle, including:
- AI Governance
- Data Privacy
- Model Risk Management
- Bias and Fairness
- Explainability
- Continuous Monitoring
- Human Oversight
- Supply-Chain Assurance
3. Emerging Tech Safeguards
- Mandates dedicated security protocols for advanced architectures:
- Generative AI
- Retrieval-Augmented Generation (RAG)
- Agentic AI
- Proactively mitigates threats such as:
- Prompt injection
- Data/model poisoning
- Misinformation
4. Human-in-the-Loop
- AI-assisted audit procedures are permitted for evidence analysis but subject to:
- Strict confidentiality controls
- Tool validation
- Competent human review
- AI outputs cannot replace an auditor's independent judgment.
5. Strategic Alignment
- Interoperable with MeitY's India AI Governance Guidelines (2025) under the IndiaAI Mission.
- Aligns with global standards like ISO/IEC 42001.
6. Seven Guiding Principles
- Security-by-Design
- Accountability and Human Oversight
- Transparency
- Fairness
- Data Privacy
- Auditability
- (Comprehensive coverage across safety and governance norms)
ISO/IEC 42001: Key Standard
- Published in 2023.
- World's first international standard for an Artificial Intelligence Management System (AIMS).
- Jointly developed by the International Organization for Standardization (ISO) and the International Electrotechnical Commission (IEC).
- Provides a certifiable framework for organizations to manage AI responsibly.
About CERT-In
- CERT-In (Indian Computer Emergency Response Team) is the national nodal agency for cyber security incidents.
- Established under Section 70B of the Information Technology Act, 2000.
- Functions under the Ministry of Electronics and Information Technology (MeitY).
- Mandate includes: incident response, vulnerability management, emergency response, and cybersecurity awareness.
Significance for India
- Governance: Moves India from principle-based AI governance to auditable, enforceable standards.
- Trust: Addresses risks of autonomous AI acting without oversight, building digital trust.
- Global Alignment: Interoperability with ISO/IEC 42001 strengthens India's position in global AI governance forums.
- Safe Innovation: Balances innovation with security-by-design, supporting the IndiaAI Mission.
- Economic: Standardized audits promote responsible adoption of AI across sectors.
Related Provisions & Concepts
- IT Act, 2000 – Section 70B: Establishes CERT-In and its powers.
- IndiaAI Mission: Umbrella initiative for India's AI ecosystem development.
- MeitY: Nodal ministry for IT and digital governance.
- Responsible AI: Focus on fairness, transparency, explainability, and human oversight.