Study guide
The PMI-CPMAI Exam: Format, Eligibility, Fees & Scoring
PMI Certified Professional in Managing AI PMI-CPMAI
Track your progress as you read
Finished sections get a check mark.
1 Chapter 1: PMI-CPMAI Exam Foundations & the CPMAI Methodology
The PMI-CPMAI Exam: Format, Eligibility, Fees & Scoring The CPMAI Methodology: Six Phases and Data-Centric Iteration CPMAI vs Waterfall, Agile, and CRISP-DM
2 Chapter 2: AI and Machine Learning Fundamentals for CPMAI
The Seven Patterns of AI Supervised, Unsupervised & Reinforcement Learning Why AI Projects Fail: Value, Data Readiness & Right-Sizing
3 Chapter 3: Domain I — Support Responsible and Trustworthy AI Efforts (15%)
Oversee the Privacy and Security Plan Manage AI/ML Transparency in Data and Algorithm Selection Conduct Bias Checks on Model, Data, and Algorithm Monitor Regulatory and Policy Compliance Manage Accountability Documentation and the Audit Trail
4 Chapter 4: Domain II — Identify Business Needs and Solutions, Part 1 (26%)
Identify the Problem to Be Solved Evaluate Initial AI Feasibility Conduct Security, Safety, and Ethics Risk Assessments Develop the AI Project Scope Statement Determine Return on Investment
5 Chapter 5: Domain II — Identify Business Needs and Solutions, Part 2
Manage Adoption and Integration Risks Draft the AI Solution Define Success Criteria and KPIs Support Business Case Creation Identify Project Resources: People, Hardware, and Contractors
6 Chapter 6: Domain III — Identify Data Needs, Part 1 (26%)
Define the Required Data Identify Data Subject-Matter Experts Identify Data Sources and Locations Coordinate the AI Workspace and Infrastructure Gather the Required Data
7 Chapter 7: Domain III — Identify Data Needs, Part 2
Check Data Privacy, Compliance, and Access Oversee Data Evaluation Determine If the Data Meets Solution Needs: The Phase 2 Go/No-Go Convey Data Understanding to Leadership
8 Chapter 8: Domain IV — Manage AI Model Development and Evaluation (16%)
Oversee AI/ML Model Technique and Algorithm Selection Oversee Model QA/QC and Configuration Management Manage AI/ML Model Training Manage Data Transformation and Data Preparation Verify Data Quality for the Data Preparation Go/No-Go Verify the Model Is Ready for Operationalization
9 Chapter 9: Domain V — Operationalize AI Solution, Part 1 (17%)
Manage Creation of the AI Solution Deployment Plan Manage the AI Solution Deployment Oversee Model Governance in Production Oversee AI Solution Metrics and Performance Monitoring
10 Chapter 10: Domain V — Operationalize AI Solution, Part 2
Prepare the Final Report and Lessons Learned Manage the AI Solution Transition Plan Oversee the AI Solution Contingency Plan
The PMI-CPMAI Exam: Format, Eligibility, Fees & Scoring
Key Takeaways Explain This Topic AI
The PMI-CPMAI exam is 120 questions (100 scored plus 20 unscored pretest) in 160 minutes with no scheduled break, preceded by a tutorial and followed by a survey of up to 15 minutes each that are not counted against the 160 minutes.
Completing the PMI-CPMAI Exam Prep Course is the only preparatory requirement, and alongside a minimum age of 18 it is the whole of PMI's eligibility list; no prior project management, technical, or AI experience or certification is required.
Results are reported as pass/fail with no numeric score, and PMI publishes neither a passing score nor a pass rate; candidates also widely report that no domain-level breakdown is provided.
Domains II (Identify Business Needs and Solutions, 26%) and III (Identify Data Needs, 26%) together make up 52% of the exam.
The bundle of prep course plus exam is USD $699 for PMI members and USD $899 for non-members before PMI's regional and membership pricing rules; the credential renews with 30 PDUs every 3 years.
Last updated: August 2026
Where the Credential Came From
PMI acquired Cognilytica , the firm that created the Cognitive Project Management in AI (CPMAI) methodology, in September 2024 . PMI kept the methodology, rebuilt the certification around it, and launched the PMI Certified Professional in Managing AI (PMI-CPMAI) on 30 September 2025 , replacing CPMAI v7 .
The blueprint behind the exam is evidence-based, and that shapes how items are written. PMI built the Examination Content Outline (ECO) from a DACUM (Developing A Curriculum) job-task analysis held in May 2025 with eight subject matter experts (SMEs) , facilitated by The Ohio State University Center on Education and Training for Employment (CETE) . That workshop identified 68 tasks professionals actually perform when managing AI work, and the validated subset of those tasks became the five exam domains. The practical consequence: scored items trace back to things a practitioner does on a real project. The exam rewards the defensible management action, not vocabulary recall.
Format and Delivery
- Element Value
- Total questions 120 — 100 scored plus 20 unscored pretest
- Time limit 160 minutes
- Breaks None scheduled
- Delivery Pearson VUE test center or OnVUE online proctored
Languages English; Arabic, Brazilian Portuguese, French, German, Japanese, Korean, Simplified and Traditional Chinese, and Spanish added from January 2026
Tutorial and survey Up to 15 minutes each, before and after the exam; neither counts against the 160 minutes
Item style Scenario-based single-best-answer items (widely reported by candidates; PMI does not publish an item-format statement)
Result Pass/fail
Three points about that table are worth holding onto. The exam is not English-only. PMI launched it in English and added Arabic, Brazilian Portuguese, French, German, Japanese, Korean, Simplified and Traditional Chinese, and Spanish from January 2026 , so a non-native English speaker has a choice to make well before booking. The clock is narrower than the appointment. A tutorial before the exam and a survey after it each run up to 15 minutes and are not deducted from the 160 minutes, so budget roughly three hours for the seat while planning your pacing against 160. You cannot identify the 20 pretest items. They are seeded among the scored ones and look identical, so answer every item as though it counts and never spend time guessing which ones are "real."
On item style, treat the description above as a candidate consensus rather than a published rule. PMI does not publish an item-format specification for PMI-CPMAI, but the ECO's task-based construction and consistent candidate reports both point the same way: long situational stems with one defensible best action. Prepare for that shape and you are prepared either way.
The Five Domains and Their Weights
- Domain Weight Scored items (approx.)
- I. Support Responsible and Trustworthy AI Efforts 15% 15
- II. Identify Business Needs and Solutions 26% 26
- III. Identify Data Needs 26% 26
- IV. Manage AI Model Development and Evaluation 16% 16
- V. Operationalize AI Solution 17% 17
Domains II and III together are 52% of the exam. More than half your scored items come from framing the business problem and working out what data the solution needs — before a model exists. Candidates arriving from a technical background routinely over-prepare Domain IV (16%) and under-prepare that front half. Budget study time against these weights, not against your comfort zone.
Eligibility, Fees, and Windows
Eligibility: you must be at least 18 years old and complete the PMI-CPMAI Exam Prep Course . Those are the only two requirements. No prior project management, technical, or AI experience is required, and no other certification — PMP included — is needed.
Fee: the bundle (prep course plus exam) is USD $699 for PMI members and USD $899 for non-members . Both figures are US dollars and are list prices — PMI applies regional pricing in some markets and its own membership pricing rules, so the amount you are actually charged can differ. Membership is not required for eligibility; it only changes the price.
Certification window: you have 12 months from purchase to obtain the certification.
Retakes: up to 3 attempts within a 365-day eligibility period . PMI suggests waiting 30 days between attempts.
Renewal: 30 PDUs (professional development units) every 3 years .
How Scoring Works, and What PMI Does Not Publish
The result is pass/fail , and you receive no numeric score . PMI publishes neither a passing score nor a pass rate for PMI-CPMAI; outcomes are determined by psychometric analysis of the scored items. Third-party sites quoting "about 70%" are estimating, and a number you cannot verify will not help you on exam day.
On feedback, be precise about what is known. PMI does not publish a policy stating that domain-level results are withheld, but candidates report consistently that the result arrives as a bare pass or fail with no per-domain breakdown. Plan on the reported behavior: assume you will learn the outcome and nothing else.
Two study consequences follow. First, you cannot reliably fail and then triage — do not count on a diagnostic report telling you that Domain III sank you, so a retake plan has to be rebuilt from the published weights and your own honest self-assessment. Second, because there is no published cut score, "strong in four domains, weak in one" is not a strategy. Walk in broadly competent, weighted by the official percentages.
Exam-Day Strategy
Pacing. 160 minutes across 120 items is 80 seconds per item . Scenario stems run long, so use checkpoints instead of watching the clock every question: roughly 30 items by minute 40, 60 by minute 80, 90 by minute 120 . That leaves about 40 minutes for the final 30 items plus review. There is no scheduled break, so plan hydration, seating, and screen fatigue in advance — an OnVUE session in particular will not let you walk away.
Read for the phase. Nearly every stem narrates a moment inside the CPMAI lifecycle: a sponsor pushing to deploy, a data owner refusing access, an evaluation that misses a KPI, a model quietly degrading in production. Before reading the options, name the phase — Business Understanding, Data Understanding, Data Preparation, Model Development, Model Evaluation, or Model Operationalization. Naming it usually eliminates one or two options on the spot, because the strongest distractors are correct actions taken from the wrong phase.
Handle "what should the project manager do FIRST." These items test sequence, not merit . Expect all four options to be things a competent PM might eventually do. The right answer is the one that must happen before the others make sense — typically validating the problem, the data, or the constraint before committing to a solution, a number, or a tool. When two options both look sensible, ask which one, if skipped, would turn the other into wasted work.
Recurring wrong-answer shapes worth recognizing on sight:
Starting with a technology, vendor, or model architecture before the business problem is confirmed.
Committing to an accuracy figure before the data has been explored.
Skipping or compressing Data Understanding to protect a schedule.
Treating deployment as a one-time handoff with no monitoring, owner, or retraining trigger.
Escalating to a sponsor before doing the analysis that would make the escalation actionable.