- What the CPMAI Certification Actually Is Today
- Exam Format and Testing Mechanics
- The Five Weighted Domains
- Domains vs. Methodology Phases vs. Course Modules
- Registration, Bundle Structure, and Attempt Rules
- Taking It Remotely or at a Test Center
- How Scoring and Feedback Work
- Who Benefits From the Credential
- A Domain-Weighted Preparation Plan
- Keeping the Certification Active
- Frequently Asked Questions
- CPMAI means Cognitive Project Management for AI; the current exam is the PMI-CPMAI, which replaced CPMAI v7 on September 30, 2025.
- The exam has 120 questions (100 scored, 20 unscored) in 160 minutes, with no scheduled breaks.
- Identify Business Needs and Solutions and Identify Data Needs are each 26%, together 52% of the blueprint.
- The 21-hour PMI-CPMAI Exam Prep Course must be completed before you can schedule the exam.
What the CPMAI Certification Actually Is Today
CPMAI stands for Cognitive Project Management for AI. It began as a methodology developed by Cognilytica for running AI and machine learning projects, and it is now administered by the Project Management Institute (PMI). The current credential is the PMI Certified Professional in Managing AI (PMI-CPMAI), introduced on September 30, 2025 to replace CPMAI v7. If you want the definitional basics first, our explainers on what CPMAI is and what CPMAI stands for cover the naming history in more depth.
The distinction matters for anyone studying from older materials. Rules, content, and formats from the legacy v7 exam and from CPMAI+E are not interchangeable with the current exam. If a study resource references the older structure, treat its details as potentially outdated and verify them against PMI's published content outline.
What makes this certification different from a conventional project management credential is its subject: it tests whether you can steer AI initiatives from business problem through data readiness, model development, and operational deployment, with responsible AI woven throughout. It does not require you to build models yourself. It requires you to manage the people, data, decisions, and risks around them.
Exam Format and Testing Mechanics
The exam is delivered by Pearson VUE, either at a participating computer-based testing center or through OnVUE remote proctoring, subject to regional availability. Here is the structure at a glance:
| Element | Detail |
|---|---|
| Total questions | 120 |
| Scored questions | 100 |
| Unscored pretest questions | 20 (not identifiable during the exam) |
| Testing time | 160 minutes |
| Scheduled breaks | None |
| Delivery | Pearson VUE test center or OnVUE remote proctoring |
| Reference materials | Closed book; built-in calculator provided |
Three details deserve emphasis. First, the 20 unscored pretest items look identical to scored ones, so you should answer every question with full effort rather than trying to guess which ones count. Second, the exam has no scheduled breaks. That rule is specific to this exam and takes precedence over what you may have read in generic handbook language about breaks on other PMI exams. Plan your hydration and logistics accordingly. Third, the tutorial and survey time sit outside the 160-minute exam timer, so that clock covers only the questions themselves.
Roughly 160 minutes for 120 questions works out to a little over a minute and a quarter per item. For a realistic read on how that pacing feels, see our difficulty guide, and for question-style repetition you can use the timed sets on the main practice test site.
The Five Weighted Domains
The exam content outline defines five weighted domains. These percentages are what actually drive the proportion of questions you will see, though exact question counts per domain can vary by exam form. The outline also describes its task enablers as illustrative rather than exhaustive, so treat the listed examples as representative, not as a complete checklist.
| Domain | Weight |
|---|---|
| Domain 1: Support Responsible and Trustworthy AI Efforts | 15% |
| Domain 2: Identify Business Needs and Solutions | 26% |
| Domain 3: Identify Data Needs | 26% |
| Domain 4: Manage AI Model Development and Evaluation | 16% |
| Domain 5: Operationalize AI Solution | 17% |
Domain 2: Identify Business Needs and Solutions (26%)
This is where AI projects succeed or fail before any data is touched. Expect scenario questions about whether AI is the right tool at all.
- Translating a stakeholder's vague ambition into a defined problem that AI can plausibly address
- Distinguishing business value and feasibility from technical novelty
- Recognizing when a simpler, non-AI solution is the better answer
- Setting success criteria and aligning stakeholders before development begins
Domain 3: Identify Data Needs (26%)
Data is the other heavyweight. The exam treats data readiness as a project management problem, not just an engineering one.
- Determining what data a use case requires and whether it exists, is accessible, and is fit for purpose
- Assessing quality, representativeness, and gaps that could undermine a model
- Understanding data governance, ownership, and acquisition constraints
- Planning data preparation effort realistically in the project schedule
Domain 1: Support Responsible and Trustworthy AI Efforts (15%)
Responsible AI is not confined to one corner of the exam; it surfaces in scenarios across all domains, even though it carries its own weight here.
- Fairness, bias, transparency, privacy, and accountability considerations
- Recognizing when a project decision creates ethical or compliance risk
- Embedding trustworthiness checks into the project lifecycle rather than bolting them on at the end
Domain 4: Manage AI Model Development and Evaluation (16%)
You are not asked to write algorithms, but you are expected to manage the iterative nature of model work.
- Overseeing model selection, training, and iteration at a management level
- Interpreting evaluation results and deciding whether a model meets the agreed success criteria
- Handling the trade-offs and uncertainty inherent in experimental work
Domain 5: Operationalize AI Solution (17%)
A model that never reaches production delivers no value, and one that degrades silently can cause harm.
- Planning deployment, integration, and handoff to operations
- Monitoring performance over time and responding to drift or changing conditions
- Managing change, adoption, and ongoing governance after go-live
For a deeper walk through each area, read our complete guide to the five CPMAI content areas.
Domains vs. Methodology Phases vs. Course Modules
One of the most common sources of confusion is conflating three different lists that all describe AI project work. The CPMAI methodology has its own phases, the 21-hour prep course is organized into its own modules, and the exam is built from the five weighted domains above. These are not the same lists, and they do not map one-to-one.
In practice, this means you can finish the course feeling comfortable with the methodology's phases and still be underprepared for the domain framing the exam uses. After the course, reorganize your notes by domain and check each against the official content outline from PMI.
Registration, Bundle Structure, and Attempt Rules
The advertised product is a bundle that combines the preparation course and the certification exam, rather than a standalone exam fee. Key mechanics:
- Course first: You must complete the 21-hour PMI-CPMAI Exam Prep Course before you can schedule and take the exam. The 21 hours is course duration, not exam time.
- No prerequisites: No prior project management, technical, or AI experience or certification is required. See our requirements breakdown for the eligibility details.
- Twelve-month window: You have 12 months from purchase to obtain your certification.
- Three total attempts: Within that eligibility year you get up to three total exam attempts, which is not the same as three retakes after a first try.
- Paid additional attempts: Further attempts require payment. The content outline recommends about 30 days of preparation before a retake, but that is a recommendation, not a mandatory 30-day waiting period.
- Reapplication wait: If you use all three attempts within the eligibility year without passing, the general PMI handbook requires a one-year wait from your last exam before reapplying.
- Membership: PMI membership is optional, and fees vary by region and membership status.
On pricing: the live issuer pages we checked displayed loading placeholders instead of current prices, so we are not quoting specific bundle, retake, or renewal amounts here. Earlier figures circulating online could not be reverified against a retrieved PMI page. Check PMI directly before budgeting, and use our cost breakdown for the components worth comparing.
Taking It Remotely or at a Test Center
Both delivery routes are available subject to regional availability, and they have different rules about what is on your desk.
| Consideration | OnVUE (Remote) | Test Center |
|---|---|---|
| Setup | System check, identity verification, private testing space | Check-in at a Pearson VUE site |
| Writing materials | Books, notes, paper, and physical writing materials prohibited unless specifically approved; digital whiteboard provided | Follow materials supplied by the administrator at the site |
| Calculator | Built-in calculator via the testing software | Built-in calculator via the testing software |
| Breaks | None scheduled | None scheduled |
Do not assume the scratch-material rules are identical across both options. Remote restrictions on paper and physical writing tools should not be confused with whatever a test center supplies. Practice using a digital whiteboard if you plan to test remotely, and run the OnVUE system check well before exam day so a webcam or network problem does not surface at the worst moment. For scheduling logistics, see our exam dates and scheduling guide.
How Scoring and Feedback Work
PMI uses a criteria-based, psychometrically established passing standard and reports a pass/fail result along with domain-level diagnostic feedback. We were not able to verify a numerical passing percentage or an official pass rate, so we do not state one, and you should be skeptical of any site that does so without citing PMI. Our pages on the passing score and pass rate data explain what is and is not publicly known.
Key Takeaway
Because feedback is reported by domain, a near-miss still tells you where to focus. If you need another attempt, use the domain-level diagnostics to target your weakest areas rather than re-reading everything equally.
Who Benefits From the Credential
The certification is aimed at people who coordinate AI work rather than only those who code it. Typical beneficiaries include:
- Project and program managers moving into AI-heavy portfolios who need a structured approach to data-centric, iterative work
- Business analysts and product owners who translate organizational problems into AI use cases
- Data and analytics leads who manage cross-functional delivery and want a shared vocabulary with business stakeholders
- Consultants and transformation leads advising clients on responsible AI adoption
Because no prior experience is required, career changers can use it as a structured entry point, though the value to employers will depend on how well you pair it with real project exposure. We do not cite specific salary numbers because none are verified for this credential; for a qualitative discussion see the earnings analysis, the CPMAI jobs overview, and our ROI assessment.
A Domain-Weighted Preparation Plan
Generic study techniques matter less here than sequencing. A sensible approach is to follow the dependency order of an AI project, then weight your review hours by domain percentage. Here is one way to sequence the weeks after you finish the prep course:
Domain 2: Business Needs and Solutions
- Drill scenarios on whether AI is appropriate and how to frame success criteria
- Reason: at 26% it is a top-weighted domain, and it sets up everything downstream
Domain 3: Data Needs
- Practice data readiness, quality, governance, and gap scenarios
- Reason: the other 26% domain; together with Week 1 you cover 52% of the blueprint
Domains 4 and 5: Development, Evaluation, Operationalization
- Work through evaluation-versus-criteria decisions, deployment planning, and monitoring
- Reason: 16% plus 17% is a combined third of the exam and builds on the earlier domains
Domain 1 and full-length review
- Revisit responsible AI themes, then sit timed mixed sets under the 160-minute limit with no break
- Reason: responsible AI appears across scenarios, so review it last to tie everything together
For a fuller methodology, our CPMAI study guide expands on pacing, and the one-page cheat sheet is useful for last-day review of facts like the exam format. If you want structured courseware options, see our overview of CPMAI training. Then pressure-test your readiness with timed questions at the CPMAI Exam Prep practice test platform.
Keeping the Certification Active
Earning the credential is not the end of the commitment. Renewal requires 30 professional development units (PDUs) every three years, and you must report them and pay the applicable renewal fee. The distribution rules are specific:
- At least 18 Education PDUs
- At most 12 Giving Back PDUs
- At least four PDUs in each Talent Triangle skill area: Ways of Working, Power Skills, and Business Acumen
Because the Talent Triangle includes Power Skills and Business Acumen alongside Ways of Working, you cannot satisfy renewal purely with technical AI content. Plan a mix early in each cycle so you are not scrambling in the final months. Current renewal fee amounts are not asserted here; confirm them on PMI's certification maintenance page.
Frequently Asked Questions
No. The current exam is the PMI-CPMAI, introduced on September 30, 2025 to replace CPMAI v7. Legacy v7 and CPMAI+E rules are not interchangeable with the current exam, so verify any study material against PMI's current content outline.
No prior project management, technical, or AI experience or certification is required. You do need to complete the 21-hour PMI-CPMAI Exam Prep Course before scheduling and taking the exam.
You have up to three total attempts within 12 months of purchase, not three retakes on top of a first attempt. Additional attempts require payment, and the outline recommends about 30 days of preparation before a retake. If you do not pass within three attempts in the eligibility year, the general handbook requires a one-year wait from your last exam before reapplying.
No. The exam has no scheduled breaks, and this exam-specific rule takes precedence over generic handbook descriptions of breaks for other PMI exams. The tutorial and survey time fall outside the 160-minute timer.
Identify Business Needs and Solutions and Identify Data Needs are each weighted at 26%, together making up 52% of the exam. Operationalize AI Solution is 17%, Manage AI Model Development and Evaluation is 16%, and Support Responsible and Trustworthy AI Efforts is 15%. Weight your study time accordingly, while remembering that responsible AI considerations can appear in scenarios across every domain.