Browser-local diagnostic
AI project readiness check
Work through twelve delivery questions and see every area that is unanswered or only partly supported. No maturity score or readiness verdict is produced.
AI project readiness check
AI project delivery
Practical project management for AI-assisted delivery
An AI-enabled project can look like ordinary digital delivery until questions about data, output quality, human responsibility, supplier change and live monitoring arrive. These resources help a project manager organise those questions without replacing specialist judgement.
One connected working system
Start with the decision in front of you
The check identifies gaps. The risk guide helps turn uncertainty into managed exposure. The governance checklist records who must answer, decide and review.
Guide and editable workbook
AI project risk management
Describe risks in cause, event and effect form, assess current and residual exposure, and plan mitigation, contingency and live review.
AI project risk guideStage-based working checklist
AI project governance checklist
Make questions, owners, evidence and review points visible from initial approval through go-live, operation and AI-assisted project work.
AI project governance checklistEditable Excel resource
AI project risk and governance workbook
Five sheets provide instructions, a blank live risk register, twelve clearly labelled example risks, a 32-question governance checklist and visible scoring lists. It contains no macros, hidden tracking or external data connections.
What needs additional attention
AI changes some project questions, not the purpose of project management
- Objective and proportionality
- Define the operational problem and a measurable outcome before the chosen technology becomes the objective.
- Data and evaluation
- Data suitability and representative tests need owners, evidence and agreed tolerances before a persuasive demonstration is treated as acceptance.
- Human responsibility
- A person named as a reviewer needs authority, competence, time, usable information and a route to override or stop the process.
- Supplier and model change
- Models, services, terms, data use and performance can change after approval, so dependencies and re-evaluation triggers must remain visible.
- Live operation
- Monitoring, incidents, fallback, change review and decommissioning need an operational owner beyond project closure.
Where the project manager's role stops
Coordinate the decisions. Do not impersonate the specialists.
A project manager can make the owner, question, evidence, dependency, decision and review date visible. Legal applicability, information security, data lawfulness, procurement approval and model fitness still need the people authorised and qualified to decide them.
Completion of a PMZ check or workbook does not establish readiness, compliance, safety, responsible AI or assurance.
Evidence and limits
Built from public guidance, translated into project work
The recurring themes are consistent with public material from the NIST AI Risk Management Framework, PMI's AI standards work, ISO/IEC 42001 and the UK government AI Playbook. PMZ does not reproduce those standards or claim formal alignment.
APM's 2026 survey of project professionals reports that AI is already being used across forecasting, decision support, risk, reporting and other project work. It is self-reported adoption evidence, not proof that the uses improve outcomes. Read the APM research.