Skip to main content
Mallory
← Back to News
cybersecurity-regulationstandards-framework-updateai-platform-security

CERT-In AI Security Guidelines Set Risk-Tiered Audits and Lifecycle Controls

Regulatory & Policy · 3 sources · Updated 4d ago · First seen Apr 28, 2022

CERT-In AI Security Guidelines Set Risk-Tiered Audits and Lifecycle Controls

India’s CERT-In has issued Artificial Intelligence Security Auditing Guidelines v1.0, establishing a risk-based framework for AI security, governance, privacy, fairness and resilience throughout the AI lifecycle. The guidelines cover organizations operating AI in India and systems processing Indian data or serving Indian users, including third-party, open-source and API-delivered models. Eight inherent-risk factors determine four tiers—Low, Moderate, High and Critical—which govern audit intensity, technical testing and oversight. Cyble’s analysis says independent audits by CERT-In-empanelled organizations are required from Tier 2 upward, with additional escalation triggers for consequential autonomous capabilities.

The framework emphasizes evidence-based control testing, supply-chain traceability, continuous monitoring and human intervention, with specific provisions for generative, retrieval-augmented, agentic and other advanced AI systems. Organizations should inventory AI deployments, assess prompt-injection exposure and reassess risk when permissions change: Cyble illustrates how allowing a support chatbot to issue refunds independently can raise it from Tier 2 to Tier 3. Its analysis also identifies qualifying AI incidents as subject to the existing six-hour reporting and 180-day log-retention requirements under CERT-In’s 2022 Directions; the supplied primary-source excerpt ends before the full incident-reporting provisions.

What this changes for youSee the reporting duties and controls this puts on the clock.
Get the context
Event timeline

How this story unfolded

Events from the most recent confirmed update back to the earliest known activity.

3 events · Apr 28, 2022
  1. Reported Nov 3, 202614d ago

    CERT-In establishes a dedicated AI cyber defence War Room

    CERT-In states that it established a dedicated AI cyber defence War Room to support coordinated cybersecurity monitoring, threat assessment, technical analysis and response.

    Cert In
  2. Reported Nov 3, 202614d ago

    CERT-In issues AI Security Auditing Guidelines v1.0

    CERT-In issued a lifecycle security and governance framework covering AI systems operated in India or processing Indian data or serving Indian users. The guidelines establish four inherent-risk tiers, require independent audits from the Moderate tier upward, and address AI-specific threats, supply-chain assurance, human oversight and continuous monitoring.

    Cert In
  3. Apr 28, 20225y ago

    CERT-In issues incident-reporting and log-retention directions

    CERT-In issued Directions under Section 70B of the IT Act requiring qualifying cybersecurity incidents to be reported within six hours and logs to be retained for 180 days. The AI auditing guidelines reference these existing requirements for reportable AI incidents.

    CERT-In Security Auditing Guideline: Does Your AI Need One?
Linked entities

Related entities

Vulnerabilities, threat actors, malware, products, organizations, and breaches Mallory has linked to this story.

Open in app →

Organizations

1 linked
Indian Computer Emergency Response Team
Do you run this?See what this changes for your reporting obligations and which controls it puts on the clock.
Get the Regulatory Context
Mallory platform

See the full picture, correlated to your attack surface.

This page covers what’s public. Mallory adds the parts that aren’t: which of your assets are affected, which threat actors are using it right now, and what to deploy next.

Exposure mappingMap indicators from this story to your assets and identify affected systems in minutes.
Threat actor evidenceEvery observed campaign, victim, and pivot linked to actors named in this story.
Associated malwareMalware, exploits, and IOCs connected to the activity described here.
Detection signaturesYARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Scheduled alertsGet matching new stories delivered to your team as they break, not the next morning.
AI threadsAsk questions about this story and take action on the answers.