PSPO-AI Assessment Preparation: A 7-Day Evidence-Based Plan

A short assessment plan works when every study session has a decision outcome. Reading AI terminology for seven days is unlikely to help if you cannot judge whether an AI-supported product action is responsible, evidence-based, and consistent with Professional Scrum.

Scrum.org’s current PSPO-AI preparation guidance identifies three test categories: AI Theory and Primer, AI Security and Ethics, and AI Product Ownership. The certification is available to learners who attend the official course.

Day 1: build a concept map, not a glossary

Explain the differences among AI, machine learning, deep learning, generative AI, and agentic AI in your own words. For each concept, write one product use and one limitation. If the explanation depends on another undefined term, simplify it.

Day 2: connect AI to Product Owner stances

Take a product problem and examine how AI might support customer understanding, vision, experimentation, backlog decisions, stakeholder communication, and delivery learning. Keep the Product Owner accountable for value and ordering decisions.

• What input evidence does the tool receive?
• What output could inform a decision?
• What must a human verify?
• Which stakeholder or customer could be affected?

Day 3: practise data and security decisions

Classify sample inputs as public, internal, confidential, personal, or restricted. Decide which information may enter an AI tool under an imagined organizational policy. Redact names only if the remaining context cannot re-identify a person or expose sensitive work.

Day 4: reason through ethics and harm

Use one case involving biased customer segmentation, inaccessible generated content, manipulative experimentation, or an unsupported product claim. Identify affected people, potential harm, evidence gaps, and the point where the team should stop or escalate.

Day 5: write evidence-aware prompts

A strong product prompt distinguishes facts, assumptions, constraints, and the requested output. Ask the model to show uncertainty, identify missing evidence, and separate source-supported statements from suggestions.

Prompt element	Example instruction
Context	Use only the supplied interview notes
Task	Group observations by customer problem
Evidence	Cite the note behind each theme
Uncertainty	Mark themes supported by fewer than three observations
Decision boundary	Do not recommend a roadmap priority

Day 6: take the Product Owner AI Open

Use the official open assessment as a diagnostic. Record why each uncertain answer was difficult. Return to the course reference material for the underlying concept instead of memorising the answer position.

Day 7: run a mixed decision review

Create three short cases: one theory question, one security or ethics decision, and one Product Ownership application. Explain each answer aloud. Your explanation should include the value goal, evidence, risk, and human accountability.

Keep a confidence log across the seven days

After each session, record one concept you can explain, one decision you still find ambiguous, and the course source you will revisit. Mark confidence only after you can apply the idea to a new scenario. This prevents familiarity with the wording from being mistaken for understanding.

On the final day, review only the ambiguous decisions and weak explanations. A short, targeted revision is more useful than rereading every page immediately before the assessment.

Use the official attempt policy carefully

Scrum.org states on the official PSPO-AI Essentials course page that the assessment requires an 85 percent passing score. Official course participants who attempt it within 14 days and do not pass receive a second attempt under the current policy. Confirm the policy attached to your class rather than relying on an old screenshot.

Download the responsible prompt and risk checklist for the daily exercises. The AgileSeekers PSPO-AI Essentials course includes official training, practical exercises, and assessment support.