In the full course
What you get
- 400 exam-style questions, each with a full explanation
- 200 flashcards, filtered by domain
- The full study guide, 24 chapters
- Timed mock exams matched to the real exam length
- A readiness score weighted by the official exam blueprint
Get full access to ISACA AAIA
All questions, timed exams, flashcards, PDF study guide download & progress tracking.
Lifetime · all courses
$29.99$19.99
One payment · future courses included
30 seconds, then straight to checkout.
Pass, or your money back
Reach 85% readiness on this course, sit the real exam, and if you don't pass we refund it in full. Applies to this single-course purchase. Terms.
More free samples
Marked, and passed
Real feedback from people who passed
“What I liked most was seeing my domain scores after each practice session. I knew exactly that I was weak on Access Control and could drill just that area. Passed CISA with a comfortable margin.”
“I failed my CISSP on the first attempt with another platform. Switched to CertCrush, focused on my weak domains using the tracking feature, and passed three months later. The explanations for wrong answers are genuinely useful, not just 'A is correct because A is correct'.”
“Honestly wasn't expecting much but this is probably the best ten bucks I've spent on exam prep. Did 20–30 questions every morning before work for 6 weeks. Passed with a comfortable margin. The timed exam mode is what really got me comfortable with the pressure.”
ISACA's Advanced in AI Audit (AAIA) is the advanced credential for IT auditors who now have to audit AI systems and use AI in their own audit work. It covers AI governance and risk, AI operations, and AI auditing tools and techniques, and you need an active CISA or another qualifying audit designation to sit it. Built for experienced auditors and advisors who assess, implement or review AI solutions.
Practice content last updated · Independently written and aligned to ISACA’s published exam objectives.
About the ISACA AAIA Exam
The ISACA Advanced in AI Audit (AAIA) is a certification for auditors who already know how to audit and now need to do it on AI. It is not an entry-level AI badge. To be eligible you must hold an active CISA or another qualifying designation, such as CIA or US CPA, so ISACA assumes you can already plan an engagement, test a control and write a finding. What the exam adds is the AI part. Can you tell whether a model's training data was fit for purpose? Can you evaluate how an organisation approves model changes, supervises outputs and decisions, and responds when an AI system misbehaves? Can you judge whether it is safe to use AI inside the audit process itself? Those are the questions AAIA puts in front of you. The content outline has three domains. AI Governance and Risk (33%) covers AI models and requirements, governance and program management, AI risk management, privacy and data governance, and the ethics, regulations and standards that apply to AI. AI Operations (46%) is the largest: data management, the AI solution lifecycle, change management, supervision of AI solutions, testing techniques, AI-specific threats and vulnerabilities, and incident response. AI Auditing Tools and Techniques (21%) covers audit planning and design, testing and sampling, evidence collection, data quality and analytics, and AI audit outputs and reports. Nearly half the exam sits in AI Operations. Auditors who revise governance frameworks and skim the operational material are preparing for the smaller part of the paper. CertCrush gives you a study guide mapped to all three domains, a practice question bank weighted to match the exam, and flashcards for the vocabulary.
Exam Domains Covered
- AI Governance and Risk33%
- AI Operations46%
- AI Auditing Tools and Techniques21%
Exam Format & Details
90 multiple-choice questions in 150 minutes (2.5 hours), which is about 100 seconds per question. Scores are reported on ISACA's 200 to 800 scale and you need 450 or higher to pass. Only correct answers count and there is no penalty for a wrong one, so never leave a question blank. The exam is computer-based and delivered by PSI, at a test centre or by remote proctoring. Registration is US$459 for ISACA members and US$599 for non-members, with a US$50 application processing fee when you apply for the certification after passing. You must hold an active CISA or another qualifying designation, such as CIA or US CPA, to be eligible. You have six months from the date you register to sit the exam. There are no performance-based questions, labs or essays. If you do not pass, ISACA allows four attempts in a rolling twelve-month period, with a 30-day wait before the second attempt and 90 days before the third and fourth.
Why Practice Questions Matter
AAIA candidates arrive with years of audit experience, and that experience can mislead. The exam asks what an auditor should evaluate, recommend or do first when the system under review is a model, and the instinct that works on a database or an ERP control is sometimes the wrong one here. Practice questions show you where your audit habits and the AI material disagree before exam day does. They also build pace. Ninety questions in 150 minutes is about 100 seconds each, and 46% of them come from AI Operations, the domain most auditors have worked in least. Working through weighted practice sets tells you early which of the three domains needs your remaining study time.
Sample Practice Questions
The following questions are a preview of the type of syllabus-aligned questions you will practise in CertCrush. They reflect the format and reasoning style of the ISACA AAIA exam, not actual exam content.
Q1.Sam wants assurance that the model running in production is the one that was approved and has not been altered since release. Which control provides this?
- A.Rate limiting on the production inference interface to prevent excessive query volumes from any single account
- B.Model signing with signature verification before the model is loaded, so any altered model gets rejected
- C.Output filtering that masks personal data in responses before they are returned to the calling application
- D.Adversarial training that includes crafted inputs in the data used to retrain the model each quarter
Domain: AI Operations
Q2.Dana's organization is moving from a predictive model built on transaction tables to a generative AI assistant trained on emails and documents. Which data risk is MOST likely to increase?
- A.Schema drift in relational tables, because unstructured sources require stricter column definitions than transaction records do
- B.Personal or copyrighted material going unnoticed, because large volumes of unstructured content are harder to inventory and check
- C.Duplicate customer rows inflating group sizes, since unstructured content is always held in a single denormalised table
- D.Loss of control totals, because unstructured sources cannot be reconciled once they enter any extract, transform, load process
Domain: AI Operations
Q3.An IS auditor reviewing AI risk monitoring finds a dashboard showing the percentage of high-risk models validated on schedule. Which description is MOST accurate?
- A.A key risk indicator, because it gives early warning that the underlying risk exposure is rising across the model portfolio
- B.A threshold, because it sets the level at which the organization must escalate a high-risk model to the governance committee
- C.A key control indicator, because it shows whether a control is working as intended
- D.A trigger, because it defines who must be told and what action they must take when a model misses its validation date
Domain: AI Governance and Risk
Q4.Which approach to auditing is MOST appropriate for a rule-based expert system?
- A.Assess it only through its training data because its behaviour is learned from examples
- B.Read the rules, test them and check who can change them
- C.Rely on probability thresholds because its outputs are always statistical estimates
- D.Treat it as a black box and judge it only on the pattern of its observed outputs
Domain: AI Governance and Risk
Q5.Sam reviews a credit model with superb development results and poor performance three months after release. The developers tested it themselves, using part of the same data used for training. What is the MOST likely cause?
- A.Overfitting that went unnoticed because validation was not done on independent data
- B.Model drift, since performance usually decays gradually over a long period after release
- C.Too little training data, which should be fixed by increasing the volume of data
- D.A weak decision threshold that management should correct by retraining immediately
Domain: AI Operations
Q6.An audit manager asks Chris to explain the difference between the annual plan and an engagement plan. Which statement is correct?
- A.The annual plan sets the objectives and criteria for each individual audit, while the engagement plan is the risk-ranked schedule that sets which audits occur and when
- B.Both documents answer the same question about how the work is done and differ only in whether the audit committee or the audit team approves them
- C.The annual plan sets which audits occur and when, while the engagement plan sets the objectives, scope, criteria, resources and program for one audit
- D.The annual plan is written after fieldwork ends to record what was done, while the engagement plan is written before any fieldwork begins
Domain: AI Auditing Tools and Techniques
Q7.Dana wants to examine a model before it goes live so that design, testing and approval controls can still be corrected cheaply. Which engagement type fits?
- A.A post-implementation review, because it tests whether the system delivered what was promised after go-live
- B.A compliance audit, because it tests the system against a single law, regulation or contract term
- C.A pre-implementation review, because its design, testing and approval controls can still be corrected
- D.A governance review, because it looks at the organization's AI strategy rather than a single system
Domain: AI Auditing Tools and Techniques
Q8.Which statement BEST describes the roles of training, validation and test sets?
- A.Training is for learning, validation is for tuning and comparing versions during development, and the test set is held back until the end for an honest estimate on unseen data
- B.Training is for tuning the model, validation is for reporting final performance to management, and the test set is used for the model to continue learning after launch
- C.Training is for the final performance estimate, validation is for the model to learn from, and the test set is used for comparing candidate versions during development
- D.Training and validation are combined into one set for learning and tuning, and the test set is reused repeatedly during development to adjust the model
Domain: AI Operations
Q9.A model performed well throughout Dana's audit period, and she gives an overall opinion that controls are effective. Which statement about that opinion is correct?
- A.It guarantees the model will not drift in later months provided the current controls continue to operate
- B.It extends to every AI system in the organization because the same controls apply to all of them
- C.It applies to the whole model lifecycle, including parts of the system that were excluded from the scope
- D.It applies to the scope, criteria and period examined and gives reasonable assurance, not a guarantee
Domain: AI Auditing Tools and Techniques
Q10.A self-review threat cannot be avoided by reassigning staff because only one auditor can do the work. Which approach BEST addresses the threat?
- A.Disclose the impairment and put safeguards in place, such as independent review of the work and disclosure in the report
- B.Proceed without comment because the auditor's good intentions remove the threat
- C.Replace the planned tests with inquiry only, since interviews are less likely to be biased by earlier involvement
- D.Ask the auditee to approve the auditor's objectivity in writing and treat the matter as closed
Domain: AI Auditing Tools and Techniques
ISACA AAIA guides & exam news
Best AI Governance Certification in 2026: AIGP vs AAIA vs AAISM vs ISO 42001 (And What the AI Act Delay Changes)
Four credentials now compete for the same AI governance job adverts, and they are not interchangeable. Here is what AIGP, ISACA AAIA, ISACA AAISM and the ISO 42001 Lead Auditor route actually cost, who can sit them, and which one matches your role.
ISACA AAIA Explained: Domains, Cost and Is It Worth It in 2026?
A full breakdown of the ISACA AAIA (Advanced in AI Audit) certification: the three exam domains and their weightings, the real cost, the CISA prerequisite, and an honest verdict on whether it is worth it in 2026.
ISACA CISA Practice Questions: 25 Exam-Style Examples With Answers and Explanations (2026)
Twenty-five CISA practice questions written to the real domain weightings, with the reasoning behind every answer. Covers all five job practice domains, plus the exam format, pass mark and the auditor mindset ISACA actually tests.
ISACA Cybersecurity Fundamentals Is Being Retired: What Replaces It and Should You Buy It Before 1 December 2026?
ISACA stops selling the Cybersecurity Fundamentals Certificate on 1 December 2026 and sunsets the programme on 1 June 2027. Here is what happens to the certificate you already hold, what the CCS replacement looks like, and whether buying the exam now is worth it.
Frequently Asked Questions
What is included in the free ISACA AAIA sample?
The free sample includes 10 syllabus-aligned practice questions, sample flashcards, and a preview chapter from the study guide. No account or payment is required to try the sample.
How many questions are in the full ISACA AAIA course?
The full ISACA AAIA course includes 400 practice questions, covering all 3 exam domains. Every question carries a full explanation for the right answer and the wrong ones.
Are these official ISACA exam questions?
No. CertCrush questions are independently written and syllabus-aligned. They mirror the format, difficulty, and reasoning style of the official exam. We are not affiliated with or endorsed by ISACA.
Which domains does the ISACA AAIA course cover?
The course covers 3 exam domains: AI Governance and Risk, AI Operations, AI Auditing Tools and Techniques.
Can I study on mobile?
Yes. CertCrush is fully responsive and works on phones, tablets, and desktops. The timed exam, flashcards, and study guide all work on mobile without installing an app.
What happens when I create an account?
Creating a free account lets you access full courses, track your weak areas by domain, and resume practice sessions across devices. No credit card is required to register.
Start with 10 free questions
No account, no card. The full ISACA AAIA course is $9.99, once.