AI Adoption Strategies
Definition
PMBOK 8 (Appendix X3) classifies opportunities to use AI in project work by task complexity and the need for human supervision — the more complex the task, the more human intervention is required for a high-quality outcome. Three categories:
- Automation — low-complexity tasks needing little human intervention in the final output. Examples: report generation, document analysis, meeting/call summarization. Standard prompts can be created and reused across projects and teams.
- Assistance — the tool complements analysis and iteratively builds toward an expected output. First iterations are not complete without further analysis and refinement; the project professional must review results for accuracy and completeness. Examples: drafting a risk register, a scheduling plan with buffers.
- Augmentation — enhancing existing capabilities and exploring new ones on strategic, complex tasks (e.g., balancing portfolio options to maximize ROI, risk forecasting based on external variables). The professional uses the tool as a brainstorming partner, exchanging ideas and refining results over multiple iterations.
Related concepts
Exam angle
- Match the level to the task: reporting/summarization = automation; drafts of registers and plans = assistance (must be reviewed); strategic forecasting/portfolio trade-offs = augmentation (iterate as a partner)
- Assistance output is never final: wrong = accept an AI-drafted risk register or plan as complete; right = review, refine, and validate before use
- Neither worship nor prohibit: PMI’s pattern is “govern and verify” — blanket bans on AI and blind acceptance of AI output are both wrong answers
- Complexity drives supervision: if a scenario emphasizes strategic stakes or unpredictability, expect more human iteration, not less