LAB / WMSI-EXP-086
Human Verification / Automation-Bias Duplication.
Does “AI analyzes, human verifies” save work, duplicate work, or induce rubber-stamping?
01 / QUESTION
Human-AI Interaction and Operational Safety
Does “AI analyzes, human verifies” save work, duplicate work, or induce rubber-stamping?
This is a protocol scaffold, not a completed experiment or a frozen execution plan. No measured effect or research conclusion is claimed here.
Hypothesis to test
The manipulation described below produces a reproducible, condition-specific change in the measured system behavior or internal state that is not explained by the ordinary controls.
Ordinary competing explanation
The observed pattern is explained by ordinary sampling variation, prompt/representation artifacts, surface lexical cues, evaluator bias, or the known product/runtime mechanics captured by the controls.
02 / PLANNED DESIGN
How we intend to test it
Compare independent-human, AI+full verification, AI+targeted evidence audit, and AI-only workflows on matched analytical tasks.
Controls
Blind ground truth; varied model reliability; time pressure manipulation.
What we would measure
Accuracy, labor time, disagreement rate, missed AI errors, confirmation bias.
Access and instruments
Limited access or proxy measurement. This describes access requirements, not verified readiness on our Spark.
Task platform; timing logs; human study protocol.
Before a full run
Feasibility=PARTIAL; priority=P1. Before confirmatory execution, run a small pilot to estimate variance/cost, freeze the scoring rules, and assign a run-count target.
Model revisions, final prompts, sample sizes, cost limits, and scoring rules must be settled through the pilot and protocol-freeze process.
03 / INTERPRETATION
What could fool us
Participant expertise imbalance, learning effects, leading instructions.
What would weaken the idea
If full verification costs no more than human-only and retains accuracy, duplication concern weakens; if targeted audit performs best, workflow design matters.
Observation, inference, speculation, unknowns, and apparatus errors will be labeled separately. One striking transcript cannot establish a general effect.
04 / RECORD
Evidence we plan to preserve
Task ground truth, AI output, human edits/checks, timing, confidence, final decision.
Planned visual evidence
Accuracy-time frontier; error acceptance matrix.
No research runs or raw datasets are published for this experiment. The September 26, 2026 registry and protocol compendium (version 1.0) are the sources for this planning record.
Back to all 90 experiments