Certified AI
Governance Auditor (CAIGA)
The how-to, hands-on AI Audit Certification
Become the auditor who can test whether AI governance works and prove it with evidence.
Gain a multi-standard AI audit methodology mapped across 90+ frameworks, including ISO/IEC 42001, the NIST AI RMF, and the EU AI Act.
Earn the Certified AI Governance Auditor credential by passing exam 612-40 and maintain it with 120 CPE credits. Qualify with 2+ years of experience in audit, GRC, risk, compliance, or security, or with an active recognized credential.
40 hours · Self-paced
ComplyX access · Run an audit on a live governance platform
180+ controls • Organized across 12 families
90+ standards & regulations · One audit method
Auditor’s Toolkit · Practical resources for the audit
Beyond checklists · Live, hands-on audit training
eCourseware
4.7 on Trustpilot
from 350,000+ trained professionals
Trusted By
Trusted by 7 of the Fortune 10, 47 of the Fortune 100, the U.S. Department of Defense and NATO.
Built for auditors. Ready for AI.
Move beyond policy checklists to evidence-based AI governance auditing.
The Certified AI Governance Auditor (CAIGA) equips internal auditors, IT auditors, GRC and compliance professionals, and AI governance practitioners with a multi-standard audit methodology supported by controls mapped to ISO/IEC 42001, the NIST AI RMF, the EU AI Act, and 90+ authoritative sources.
Learn to use ComplyX to assess AI governance controls, evaluate supporting evidence, identify gaps, and deliver defensible audit findings.
Why CAIGA
The industry’s first hands-on AI governance audit certification
Build a repeatable audit method, use the Auditor’s Toolkit, and complete a real-world AI governance audit.
the lab
Test what frameworks only describe
Turn AI governance requirements into audit tests
Knowing ISO/IEC 42001, the NIST AI RMF, and other frameworks is different from knowing how to conduct an audit. CAIGA teaches you how to perform the audit, not just interpret the framework.
Run an AI governance audit in ComplyX
Using ComplyX lab access, you audit Meridian Group, a realistic case enterprise. ComplyX provides hands-on AI governance audit practice through trial access, with 180+ controls across 12 families mapped to 90+ frameworks, including ISO/IEC 42001, the NIST AI RMF, and the EU AI Act.
Practice testing controls, evaluating evidence, and identifying governance gaps through guided labs and the Auditor’s Toolkit, then demonstrate your audit knowledge through an open-book, proctored exam.
Meridian Group is a fictional organization create by EC-Council form educational and assessment purposes. It is not affiliated with or intended to represent any actual organization with the same or a similar name.
ComplyX provides hands-on AI governance audit practice through trial access, with 180+ controls across 12 families mapped to 90+ frameworks, including ISO/IEC 42001, the NIST AI RMF, and the EU AI Act.
Description of the AI assistant goes here.
the method
Apply one audit methodology across 90+ frameworks
CAIGA uses a common, control-based audit approach that applies across more than 90 AI governance frameworks, standards, and regulations. You learn to assess the control, rather than memorize individual frameworks.
Evaluate AI governance through control-by-control testing
Work at the control level to learn what the control requires, what evidence to look for, and how to test it. This turns governance knowledge into practical audit execution.
Pass an exam that tests judgment, not memorization
The exam is scenario-based and focuses on applying audit principles instead of recalling clauses or definitions. It tests whether you can make sound audit judgments in realistic situations.
Keep the courseware as your audit manual
The courseware is built to remain useful after certification as a practical reference for deciding what to test, what evidence to examine, and how to approach individual controls during an audit engagement.
Work with the Auditor’s Toolkit
Use the Auditor’s Toolkit alongside ComplyX, guided audit scenarios, and the lab guide to plan tests, evaluate evidence, and document findings.
Lead the ISO/IEC 42001 internal audit
Organizations pursuing ISO/IEC 42001 certification must audit their AI management systems at planned intervals, and certification bodies expect to see evidence of that audit program. CAIGA gives internal auditors the method and control-level skills to conduct the audit.
Audit AI against AI-specific controls, not an IT checklist
Audit teams without AI skills are left with two poor options: leave AI out of the audit plan or audit it with IT checklists that may not address the right controls. CAIGA teaches which controls apply to AI systems and how to test them.
Deliver independent assurance, inside or outside the organization
Teams that build AI governance programs cannot independently assure their own work. CAIGA trains auditors to test that work, whether they operate within an internal audit function or assess clients and vendors externally.
What you get
Course, exam, and control library
20-module eCourseware
125 topics. Short lessons built around control objectives and evidence requirements.
Exam 612-40
120 scenario-based questions. 4 hours. Open book, proctored, and scheduled through the EC-Council Exam Portal after enrollment.
AI Unified Controls Framework (AI UCF)
The AI UCF is the 180+ control library within ADG (Adopt. Defend. Govern.), EC-Council’s AI governance operating system. The AI UCF includes 180+ controls across 12 families, mapped to 90+ frameworks, standards, and regulations.
The library
180+ controls. 12 families. 90+ frameworks.
180+
external standards
90+
720+
20
modules, 125 topics, 40 hours
12
audit domains spanning design through review
The Certification
Certified AI Governance Auditor
This certification is awarded in recognition of successfully completing all requirements for the Certified AI Governance Auditor (CAIGA) credential and passing an examination administered by EC-Council.
The problem
Continuous AI governance does not
stop at a border.
Many frameworks, one set of controls
What CAIGA validates
Career opportunity
Built for professionals
validating continuous AI governance
GRC & Risk
GRC, risk, and compliance professionals who must test AI governance claims against evidence, not policy alone, including AI & Enterprise Risk Managers who turn AI risk inventories and third-party AI risk into controls they can test and report on.
Internal Audit
Internal auditors and IT audit managers who extend traditional audit skills into AI-specific evidence and control testing.
ISO/IEC 42001 Track
ISO/IEC 42001 implementers and certification-track auditors who build or assess an AI management system that is audit-ready from the start, and seek a multi-standard AI audit methodology.
Third-Party Risk
Vendor-assessment professionals who evaluate AI vendors and supply-chain governance.
Security & Privacy
Security and privacy practitioners and auditors who build, deploy, test, secure AI systems, respond to AI incidents, and verify that AI-specific controls are actually enforced, including those who need evidence that holds up across the EU AI Act and other regulations.
Governance Leadership
CISOs, data protection officers, and governance leaders who build or sponsor an internal AI audit function.
The operating system for
responsible AI
Adopt. Defend. Govern.
ADG turns AI governance principles into a unified control framework mapped to the NIST AI RMF, ISO/IEC 42001, the EU AI Act, and other evolving requirements.
CAIGA teaches auditors to test those controls against evidence and assess whether governance is working as intended. ADG provides the control foundation; CAIGA provides the audit methodology.
ADG connects
- 90+ standards & regulations
- 4 industry threat catalogs
- 30/60/90-day roadmap
- 9 attack surfaces
- ISO/IEC 42001 NIST AI RMF
The global AI governance framework landscape
90+ named standards.
One AI unified controls
framework.
A CAIGA-certified auditor is trained to assess AI governance across 90+ regulations, frameworks, and standards, rather than working within a single framework.
Global enterprises operate across multiple jurisdictions, regulatory requirements, and governance frameworks. CAIGA prepares auditors to assess AI governance consistently across borders, frameworks, and jurisdictions. C|AIGA is not affiliated with, accredited by, or endorsed by ISO, IEC, NIST, OECD, CSA, MITRE, OWASP, MAS, IMDA, or any other standards body or regulatory authority.
NIST AI RMF
144
EU AI Act
141
OECD AI Principles
140
ISO/IEC 42001
140
CSA AI Controls Matrix (AICM)
96
AIUC-1
93
Google SAIF
78
MITRE ATLAS
73
OWASP Top 10 for LLM Applications
69
Monetary Authority of Singapore (MAS) Guidelines on AI Risk Management
57
ISO/IEC 27001:2022
57
Infocomm Media Development Authority (IMDA) Model AI Governance Framework
55
20 comprehensive modules
Program overview
Master multi-standard AI governance auditing across the enterprise. CAIGA covers the full lifecycle, from design through review.
Module 01
Foundations of AI Governance Auditing
What You'll Learn
- Differentiate the audit assumptions that hold for deterministic IT systems from those that fail when applied to probabilistic, non-deterministic AI systems.
- Deconstruct the AI ontology into its three constituent levels (model, AI system, and AI agent) and correctly classify a given production artifact against each level.
- Classify AI-specific harm into technical/security, fairness/bias, and operational/emergent categories, and identify which evidence type tests for each.
- Evaluate a set of existing IT audit controls (access management, change control, a SOC 2 Type II report) for what they do and do not cover when applied to an AI system.
- Explain the structure of the CAIGA curriculum and certification exam well enough to plan a personal study sequence and recognize how these foundational concepts recur in later technical domains.
Module 02
The Global AI Governance Framework Landscape
Map the global AI regulatory and standards landscape, and distinguish binding requirements from voluntary guidance before testing a single control.
What You'll Learn
- Legally binding regulation versus voluntary standards, and how each is actually enforced
- The NIST AI RMF’s four functions (Govern, Map, Measure, Manage) and how each depends on the others
- ISO/IEC 42001 as a certifiable AI management system, including Annex A control groups and companion standards
- EU AI Act risk tiers, regional instruments across the Americas, Europe, India, the Middle East and APAC, and industry baselines including OWASP, MITRE ATLAS and Google SAIF
AI/ML & MLOps Fundamentals for Auditors
Build the technical literacy needed to read AI evidence and challenge it effectively, with enough depth to test the system, not build it.
What You'll Learn
- Major AI/ML categories: supervised, unsupervised and reinforcement learning; classical ML, deep learning, generative, retrieval-augmented and agentic architectures
- The end-to-end MLOps pipeline and the governance-relevant checkpoint at each stage
- Generative AI and RAG concepts (tokens, embeddings, context windows and vector databases) and where controls must attach
- How to evaluate model cards, architecture diagrams, logs and API specifications for what is missing, not only whether they exist
AI System Design & Requirements Governance
What You'll Learn
- Why design-stage governance underpins the whole program, and the failures weak controls allow downstream
- How to sample each design-stage control, what evidence corroborates it, and what to ask in interviews
- Testing whether a declared autonomy and agency boundary is actually enforced technically, using design-time and runtime evidence
- Assessing impact assessments and prohibited-practice screening records for genuineness rather than retrofit
AI Data Governance & Protection
What You'll Learn
- Access control and personal-information protection for AI training and operational data
- Data quality, provenance, retention, integrity and lineage across the full data lifecycle
- Cross-border data transfer risk assessments and data-subject rights fulfilment processes
- Training-data bias detection, intellectual-property controls, and AI sustainability and ESG reporting evidence
Module 06
AI Build, Integration & Secure Development
What You'll Learn
- Identity and access management applied to AI pipelines, including non-human and agent identities
- Model lifecycle controls spanning supply-chain risk, poisoning detection, value alignment and build-time action oversight
- Model integrity, versioning, provenance and decommissioning, and the behavioural baseline runtime monitoring depends on
- Adversarial and platform security: model extraction, prompt leaking, RAG and vector-database security, API and endpoint management
AI Deployment & Change Management
What You'll Learn
- Scoping a deployment audit across platform engineering, DevSecOps and the AI Governance Council
- Testing release criteria, evaluation thresholds and separation between requester and approver
- Staged rollout and rollback: halt criteria, timed rollback exercises, and full restore of model, prompts, configuration, retrieval index and feature pipelines
- Tracing a sampled release through the gate and grading it as a major or minor nonconformity or an opportunity for improvement
Module 08
AI Evaluation, Testing & Red Teaming
Evaluate the testing evidence that proves an AI system behaves as claimed and judge whether the tester was genuinely independent.
What You'll Learn
- Risk-based testing methodology and selecting metrics appropriate to the harm being tested
- Evaluating red-teaming and robustness testing evidence
- Fairness and bias evaluation methodology, including protected-attribute testing
- When independent algorithmic bias audits are required, how to test for genuine independence, and how to evaluate pre-deployment explainability evidence
AI Runtime Security, Defense & Monitoring
What You'll Learn
- Prompt-injection and jailbreak defences, and model integrity controls for toxic output, hallucination and harmful code generation
- Drift and anomaly detection capability, and monitoring and logging coverage at scale
- Platform-level adversarial defences, including context trust ordering and denial-of-wallet protection
- Agentic AI security: trust-tiered tool registries, inter-agent communication security, and routine tested control-transfer between AI systems and human operators
AI Incident Response & Forensics
Test whether an organization can detect, contain, investigate and report an AI incident, including one caused by an agent acting autonomously.
What You'll Learn
- How an AI-specific incident differs from a conventional security incident
- Forensic evidence collection and containment mechanisms for agentic systems
- Integration between AI incident response and enterprise business continuity and disaster recovery
- Compliance with statutory content-takedown and notice-and-action obligations
AI Accountability, Roles & Organizational Governance
Trace every consequential AI decision to a named owner with authority to act and treat an untraceable one as a finding.
What You'll Learn
- Governing body accountability under ISO/IEC 38507, and testing the minimum documentation standard separately from named legal ownership
- Tracing a consequential decision and a legal obligation to a designated owner using inventory, RACI, designation records, change logs, and HR records
- Role-based competence requirements and formal appointments, without accepting either inferred from a job title
- Testing governance council decisions against charter, quorum, attendance, and escalation timing and evaluating the independence of each line of defense
AI Risk Inventory & Classification
What You'll Learn
- Testing inventory completeness by reconciling against independent sources and calculating a completeness rate
- Evaluating a risk classification and tiering model against its required dimensions, and confirming the assigned tier changes treatment
- Tracing what the Statement of Applicability includes and excludes back to signed risk treatment decisions
- Lifecycle risk reassessment: sampling a real material change and tracing each re-tiering trigger through to updated controls
Module 13
AI Transparency, Explainability & Stakeholder Communication
Audit what an organization tells users, regulators and affected communities about its AI and whether any of them can understand it.
What You'll Learn
- The three audiences AI transparency obligations serve
- User disclosure, adverse-impact reporting, data transparency and affected-community engagement practices
- Explainability at scale, including decision-justification frameworks
- Content provenance and watermarking controls, emerging transparency mandates, and the accessibility and comprehension of disclosure materials
AI Oversight, Continuous Assurance & Audit Management
What You'll Learn
- Internal audit and compliance management practices for AI systems
- AI cost and FinOps governance controls
- Governance and policy management, regulatory horizon scanning, and continuous monitoring and regulatory intelligence distribution
- Distinguishing key performance indicators from key risk indicators, and evaluating multi-framework compliance validation and requirement mapping
Module 15
AI Third-Party & Supply Chain Risk Management
What You'll Learn
- Classifying AI capabilities as built, integrated or consumed, and checking the shared-responsibility matrix names specific obligations, parties and contract clauses
- Testing whether usage limits on downstream users are communicated and enforced, not only published
- Supply-chain risk management and GenAI vendor governance across due diligence, ongoing monitoring and exit
- Testing service-level commitments against independently measured data, and distinguishing independent vendor validation from renewed self-attestation
Multi-Standard Audit Methodology & Certification Practice
Master the audit discipline itself: scoping, evidence sufficiency, weighted scoring, and the certification decision that has to survive challenge.
What You'll Learn
- Calibrating a weighted, evidence-anchored maturity model with critical-control gating, so one catastrophic failure cannot be averaged into an acceptable score
- Formulating a defensible, standard-agnostic scope with the assumptions, exclusions, and boundary conditions that protect auditor and client
- Appraising digital evidence against a hierarchy of truth and a risk-stratified sampling methodology
- Grading major nonconformities using the condition-criteria-cause-consequence-corrective action structure, and architecting Stage 1 to Stage 5 engagements with structural separation of fieldwork, certification decisions and appeals
Module 17
AI Bill of Materials & Surveillance Audits
Audit the AI-BOM as a live operational asset and run the surveillance cycle that keeps a certification honest between full audits.
What You'll Learn
- What an AI Bill of Materials documents, why it exists, and how it differs from a conventional software bill of materials
- Assessing an AI-BOM for completeness, currency, and real operational value when a new vulnerability is disclosed
- Where a surveillance audit sits in the certification lifecycle and designing a risk-based surveillance sampling plan
- Triggers for unscheduled audits and determining the correct outcome: continuation, conditional continuation, suspension, or withdrawal
AI-Assisted Auditing Techniques
Use AI inside the audit itself, with the verification, independence and documentation controls that its use demands.
What You'll Learn
- Classifying an audit task using the five AI capability primitives and predicting the verification work that classification requires
- Understanding the typical error pattern of AI in audit work (visible precision failures versus invisible recall failures) and seeded recall testing
- Applying a defensible data-handling tier model that governs which class of AI tool may see which client data
- Resolving independence, automation-bias, and self-review threats, and documenting AI-assisted procedures to a standard that survives regulatory scrutiny
Converting an Operating Model into an Operating System
What You'll Learn
- What makes a governance framework a continuous operating system rather than a static operating model
- ADG’s control taxonomy: 12 control families, Entity-Owned, AI-System-Owned, or Both applicability tagging, and critical-control designation
- ADG certification mechanics, and how Entity Governance and Portfolio Composition Scores combine and are gated to produce a published rating
- A standards-neutral, four-question framework for evaluating any execution substrate, and reconciling a system-reported score against underlying evidence
Professional Ethics & Exam Preparation
Hold the line on independence and ethics under real fieldwork pressure, and prepare for the CAIGA exam.
What You'll Learn
- Upholding the CAIGA Code of Professional Ethics (integrity, objectivity, competence, confidentiality and professional behavior) under realistic pressure
- Evaluating an engagement for conflict-of-interest and independence violations, including the self-assessment ban and cooling-off requirements
- Resolving a live ethical conflict through the correct recusal and escalation procedure
- Building a CPE and recertification strategy, and synthesizing the 20-module curriculum into a timed exam-day strategy for scenario-based multi-standard questions
Career opportunity
Industries that need
AI governance auditors
All
Sectors
Any sector using AI in decisions affecting customers, patients, or citizens, including finance, healthcare, government, technology, retail, and public policy.
DCWF
Mapped to named DoD Cyber Workforce Framework (DCWF) work roles across governance, defense, and related pathways.
120
CPE credits
Career pathways
aligned to DCWF Roles
01
ISO/IEC 42001 Lead Auditor
SECURITY CONTROL ASSESSOR
Conducts comprehensive, independent testing of management systems against defined standards.
02
AI Conformity Assessment Specialist
SECURE SOFTWARE ASSESSOR
Analyzes software and system architectures to ensure compliance with predefined security and conformity requirements.
AI Risk Manager
AUTHORIZING OFFICIAL / DESIGNATED REPRESENTATIVE
Assumes ultimate responsibility for evaluating and accepting operational risk for deployed systems.
AI Compliance Officer
INFORMATION SYSTEMS SECURITY OFFICER (ISSO)
Ensures systems are operated, maintained, and disposed of in alignment with internal security and compliance policies.
05
AI Ethics & Privacy Manager
PRIVACY COMPLIANCE MANAGER
Ensures organizational adherence to privacy mandates, data subject rights, and ethical data usage.
06
Compliance Counsel (AI Regulation)
CYBER LEGAL ADVISOR
Provides legal and regulatory counsel regarding cyber, technology operations, and statutory mandates.
07
AI Security Architect
SECURITY ARCHITECT
Designs secure system architectures and integrates security controls throughout the AI deployment lifecycle.
08
AI Trust & Safety Lead
SYSTEM SECURITY ENGINEER (SSE)
Engineers and enforces runtime guardrails, system safety boundaries, and operational integrity.
Validate multi-standard audit
competence in one exam
612-40
4 hours
70%
What changes when you earn it
Stop guessing when the
framework changes.
Proof, not a policy deck
Learn the control once
The same control can appear under different names across standards. CAIGA teaches the substance, then shows you how each regime expresses it.
Stay composed when the standard changes
Complements single-standard credentials
CAIGA PROGRAM FAQs
Frequently
asked
questions
What is the Certified AI Governance Auditor (CAIGA) and who is it for?
CAIGA is EC-Council’s practitioner-level credential for auditing AI systems and AI governance programs. It is built for GRC, risk, and compliance professionals, internal auditors, third-party risk assessors, and AI governance practitioners who need to test AI governance claims against evidence rather than accept them on faith.
How is CAIGA different from an ISO/IEC 42001 Lead Auditor course?
CAIGA is not a lead auditor accreditation and does not grant authority to issue ISO/IEC 42001 conformity certificates. It teaches a cross-standard audit methodology, evidence requirements, and control mapping that apply across ISO/IEC 42001, the NIST AI RMF, the EU AI Act, and several others frameworks. Certification does not guarantee employment.
Does CAIGA replace or complement credentials like PECB's ISO/IEC 42001 Lead Auditor, ISACA's AAISM, or IAPP's AIGP?
It complements them. Those credentials go deep on one standard or one governance discipline. CAIGA fills the audit-execution gap across all of them.
What is the eligibility requirement for CAIGA certification?
For giving CAIGA exam, candidates must be at least 18 years old and qualify through one of the following two routes:
Official Training Route: Complete the official CAIGA training through the EC-Council portal and have at least two years of relevant work experience.
Eligibility Route: Hold an active recognized credential or have at least two years of relevant work experience.
The eligibility application fee is USD 100 and is non-refundable.
Recognized credentials: CISA, CISM, CRISC, CGEIT, CIA, CPA (any jurisdiction), ACCA, CA, CISSP, CCISO, ISO/IEC 27001 Lead Auditor, or ISO/IEC 42001 Lead Auditor.
Relevant experience: Audit, assurance, GRC, risk, compliance, privacy, or information security
Does the certification expire, and how many CPE credits are required to maintain it?
CAIGA certification is maintained through EC-Council’s Continuing Education (ECE) framework. Professionals must earn 120 CPE credits and pay the USD 250 continuing education fee to maintain the certification.
How does CAIGA relate to ADG and EC-Council’s other AI certifications?
ADG (Adopt. Defend. Govern.) is EC-Council’s freely available AI governance operating system. At its core is the AI Unified Controls Framework (AI UCF), a library of 180+ controls that CAIGA teaches auditors to assess.
CAIPM, COASP, and CRAGE help professionals build, secure, and govern AI using ADG. CAIGA focuses on independent assurance, teaching auditors to test those controls against evidence and verify that governance has been implemented as intended.
Is 24/7 learner support available for CAIGA certification?
Yes. Certified AI Governance Auditor (CAIGA) learners have access to 24/7 support for training access, course resources, the learning platform, and certification-related queries.
How is CAIGA different from other AI audit certifications?
CAIGA differs in three key areas: access, scope, and audit methodology.
CAIGA is open enrollment. We recommend at least one year of experience in IT audit, GRC, or compliance, or CRAGE certification, but there is no gating credential.
CAIGA covers 12 AI governance control domains across 20 modules and 125 topics, spanning the full lifecycle from design and data governance to evaluation, red teaming, runtime security, incident forensics, and third-party assurance.
The audit methodology is central to the program. Learners follow a repeatable lifecycle: Plan → Scope → Test → Evidence → Score → Report
CAIGA is also supported by a library of 180+ controls mapped to 90+ standards, with 720+ checkable audit requirements, giving auditors a practical reference beyond the exam.
Does CAIGA teach me to implement compliance or audit it?
CAIGA teaches you to audit. You test controls, document nonconformities with the corrective action required, and assess whether they have been closed. On the other hand, building and running a governance program can be learnt from the CRAGE program. Keeping these roles separate helps preserve independent assurance.
Ready to Audit AI Governance Across Borders?
Pricing and enrollment to be confirmed at launch. Request the brochure now, or have a training advisor walk you through the program.
Program Inquiry
Download Brochure
Inquire About CAIGA Training