5.3.6. Redesigning Roles for Human-AI Collaboration
💡 First Principle: AI changes tasks before it changes jobs. Most roles are bundles of tasks, some of which AI performs well (high-volume pattern recognition, drafting, summarizing, routine decisions) and some of which humans perform better (judgment in ambiguous situations, empathy, creativity, accountability, relationship-building). Role redesign rebundles work so that each does what it does best, and moves people from doing manual operations to overseeing and collaborating with AI.
From manual operations to oversight and collaboration. Typical shifts:
- A claims processor moves from reading every document to reviewing AI summaries, handling complex or disputed claims, and auditing AI recommendations.
- A customer-service agent moves from answering routine questions to resolving escalated cases, handling sensitive conversations, and coaching the AI through feedback.
- An analyst moves from building routine reports to interpreting AI-generated insights, testing assumptions, and advising decision-makers.
- New roles appear: AI product owners, AI quality reviewers, prompt and knowledge-base curators, AI governance specialists.
Balancing human strengths with AI capabilities:
| Human strengths (named in the exam guide) | AI strengths | Design implication |
|---|---|---|
| Critical thinking: judging novel or ambiguous situations, questioning outputs | Speed and consistency on well-defined patterns | Humans review edge cases and validate AI outputs |
| Empathy: understanding emotion, building trust, handling sensitive situations | Availability and scale for routine interactions | Humans handle distressed customers, complaints, and relationship moments |
| Creativity: framing new problems, original ideas, strategic choices | Generating many options and drafts quickly | AI drafts options; humans choose, refine, and own the result |
| Accountability and ethical judgment | Tireless monitoring and detection | Humans own decisions with significant impact (4.1.4) |
Strategic decisions about roles should consider business value, the risk profile of each task (where human oversight is required), employee development, and organizational values. Organizations that frame AI primarily as headcount reduction often lose the institutional knowledge and trust they need to make AI work, while those that redesign roles and invest in reskilling capture both efficiency and capability. That does not mean workforce size never changes; it means role design is the strategic lens, and any workforce changes are handled transparently and responsibly.
⚠️ Exam Trap: When asked how to approach workforce impact, prefer options that redesign roles so people shift from manual tasks to oversight, exception handling, and collaboration with AI, balancing human critical thinking, empathy, and creativity with AI capabilities, supported by reskilling. Options framing AI solely as a replacement for people, or ignoring role changes entirely, are distractors.
Reflection Question: An AI system can now draft 80% of routine underwriting decisions. How would you redesign the underwriter role, which tasks should remain human, and what new skills would underwriters need?
