Human-AI Delegation Simulator

Act as a critical learning-engineering reviewer. Review the prototype brief below as a thin-slice experiment, not as a finished product.

Return:
1. The strongest design assumption.
2. The most important plausible failure mode.
3. The highest-value falsification test for the next cycle.
4. One credible non-AI alternative that could address the same capability gap.
5. The next smallest prototype worth building.
6. What should be measured to distinguish engagement from learning, transfer, and real performance.
7. Human-agency, equity, accessibility, privacy, or governance concerns that should change the design.

Be concrete. Separate evidence-backed claims from hypotheses. Do not reward novelty for its own sake, and do not assume more AI is better.

COLLECTION: LENS Canonical
TITLE: Human-AI Delegation Simulator

CAPABILITY GAP:
When AI enters a workflow, the capability requirement changes: people must decide when to rely, when to verify, when to override, and when to revoke delegation. Accuracy alone does not establish that the human-AI team is safe or effective.

FIRST PROTOTYPE:
Create a scenario simulator in which AI advice varies in confidence, quality, and context. Learners practice delegation, verification, override, and escalation while the system records not only outcomes but the reasoning behind reliance decisions.

REPRESENTATIVE USE CASE:
An analyst receives AI-ranked cases with explanations of uneven quality. Some recommendations are right for the wrong reason; others are uncertain. The learner must decide what to accept, inspect, or escalate before seeing consequences.

LEARNING / HUMAN-SYSTEM FRAME:
Human agency is an engineered property of the team. The design should make authority, uncertainty, and correction visible, and test calibrated reliance rather than simple compliance or distrust.

QUESTIONS ALREADY IDENTIFIED:
1. Which decisions may be delegated and under what conditions?
2. Can the human detect when the AI is outside its competence?
3. Does practice improve calibrated reliance without increasing unnecessary workload?

FIRST-CYCLE FRAMING:
Understand: Define the shared task, stakes, authority boundaries, and failure modes of the human-AI team.
Map: Map information flow, AI capabilities and limits, human expertise, time pressure, accountability, override paths, and recovery mechanisms.
Instrument: Measure decision quality, calibration of reliance, verification behavior, override appropriateness, recovery from AI error, workload, and performance when the AI is unavailable.