AI-Driven Management Professional

Domains and learning objectives

A domain’s weighting is its share of the discipline, and the share of the examination paper drawn from that domain. The study material and the examination questions are written to the objectives.

1. Domain 1

   AI Capabilities and Limits

   20%

   • 1.1Distinguish generative AI, predictive analytics and rule-based automation by the type of problem each is suited to
   • 1.2Interpret the probabilistic nature of large language model output and its consequences for reliability, consistency and factual accuracy
   • 1.3Identify task characteristics that predict strong AI performance, such as pattern-rich, language-heavy and tolerance-for-review tasks, versus characteristics that predict failure
   • 1.4Evaluate claims made about an AI tool by asking what evidence, evaluation method or benchmark would substantiate them
   • 1.5Select an appropriate response to hallucinated, biased or outdated AI output in a work context

2. Domain 2

   Use Case Selection and Scoping

   22%

   • 2.1Evaluate candidate use cases using a value, feasibility, data and risk screen, and rank them for a pilot portfolio
   • 2.2Distinguish between tasks to automate, tasks to assist and tasks to leave fully human, based on error cost and review capacity
   • 2.3Calculate simple expected benefit figures, such as hours saved per week and cost per output, using spreadsheet-level arithmetic
   • 2.4Select success criteria and baseline measures for a pilot before it starts, including quality as well as speed measures
   • 2.5Identify scoping failures such as unbounded ambition, missing data, no owner and no exit criteria, and choose the correction
   • 2.6Decide among build, buy, use-existing-tools and wait options for a given scenario

3. Domain 3

   Leading Adoption and Change

   18%

   • 3.1Select communication and framing approaches that reduce fear and build realistic expectations when introducing AI to a team
   • 3.2Distinguish causes of resistance, such as job insecurity, quality distrust and skill anxiety, and match each to an effective response
   • 3.3Evaluate skill-building options for a team, from structured practice sessions to champion networks and shared prompt libraries
   • 3.4Interpret adoption signals, including usage patterns, shadow use of unapproved tools and quiet abandonment, and choose the correct intervention
   • 3.5Select what to report to senior stakeholders about an AI initiative, balancing candour about limits with credible progress

4. Domain 4

   AI-Assisted Teams and Workflows

   20%

   • 4.1Select the appropriate human review point for an AI-assisted workflow given the error cost and volume of the task
   • 4.2Distinguish accountability for AI-assisted work: who owns the output, who owns the error, and what the manager must make explicit
   • 4.3Evaluate quality control approaches for AI-assisted output, including sampling, second review and checklist verification
   • 4.4Interpret changes in team workload, skill mix and role content as AI absorbs routine tasks, and select fair responses
   • 4.5Calculate simple throughput and review-capacity figures to judge whether a proposed AI-assisted workflow is sustainable
   • 4.6Select delegation practices for AI-assisted work, including when to require disclosure that AI was used

5. Domain 5

   Governance, Risk and Responsible Use

   20%

   • 5.1Distinguish AI-related risks a manager owns, such as data leakage, confidentiality breaches and unreviewed errors, from risks owned at organisational level
   • 5.2Select the correct handling of personal, confidential or client data when staff use AI tools, applying data protection principles at decision level
   • 5.3Evaluate an acceptable-use rule set for a team, identifying gaps, overreach and unenforceable provisions
   • 5.4Identify situations that require escalation beyond the manager, including high-stakes automated decisions, regulated-sector constraints and suspected harmful output
   • 5.5Interpret the manager-level implications of emerging AI regulation and risk-tiered approaches, without reliance on any single jurisdiction's text
   • 5.6Select proportionate responses to an AI-related incident, including containment, disclosure and process correction