Use AI to expand research without treating fluent text, repeated output or model confidence as verification.
Open situation →Reasoning Practice · Advanced
The evidence comes from a related workflow
A review reports several cognitive biases across AI-assisted medical decisions. A team wants to say those biases are established in its own specialist workflow, even though the review contains almost no direct studies from that workflow.
Your decision
How should the team use the review?
Show the best first move
B. Separate direct-domain evidence from related-domain evidence; use the latter to define checks and hypotheses, not to claim the same effect is already established.
Why this move helps
Related-domain evidence can tell you what to test without proving that the same effect occurs in a different task, population or workflow. The useful question is evidence distance, not whether the neighboring field sounds similar.
Evidence still missing
- Direct studies from the target workflow.
- Whether the AI role, task, users and decision stakes match the source studies.
- A local test or monitoring design for the proposed failure mode.
Evidence note
A 2026 structured review found 12 reported biases across AI-assisted medical decision making, while only one included study came directly from pathology. That makes the review useful for generating checks in pathology, but not proof that all 12 effects are established there. This scenario is mainly about evidence transfer; Automation Bias is only a supporting lens when automated output is relied on without adequate verification.
Next action
Mark each claim as direct, transferred or hypothetical evidence before turning it into guidance.
Best first lens
Automation Bias
Use the lens to ask a better question. It does not prove that a person or team has a cognitive bias.
Other defensible lenses
The situation may have more than one explanation.
No alternative lens is required for this exercise.
Continue the path
Situation → skill → technique → action.
You can use an AI system to expand reasoning while keeping important judgments independently checkable.
Build this skill →Separate independent evidence from repetition and generated summaries.
Run the technique →