Protocol
Write acceptance criteria before asking AI to generate
If the rubric arrives after the answer, the answer gets to write part of the rubric.
When it fits
- AI can produce many plausible versions and the team tends to judge them by polish after the fact.
When to avoid it
- Do not over-specify creative work until every useful alternative is squeezed out; criteria should protect the job, not freeze wording.
Why it matters
Before generation, state the few conditions the output must satisfy: factual boundaries, required fields, audience need, constraints and observable completion. Generate against those criteria, then review the result criterion by criterion. This keeps integration and verification tied to the task rather than to surface fluency.
Steps
- The final review can reject a polished answer for a specific failed criterion without inventing standards after reading it.
An example
Before asking AI for an incident summary, require confirmed facts only, explicit unknowns, current impact, next action and no invented restoration time.
Check your result
The final review can reject a polished answer for a specific failed criterion without inventing standards after reading it.
Keep this limit in mind
- Do not over-specify creative work until every useful alternative is squeezed out; criteria should protect the job, not freeze wording.
Evidence and sources
Knowledge workers in the CHI 2025 study described critical AI-assisted work as shifting toward information verification, response integration and task stewardship.
These themes describe reported practice and should not be treated as an exhaustive model of good AI use.
The Impact of Generative AI on Critical Thinking: Self-Reported Reductions in Cognitive Effort and Confidence Effects From a Survey of Knowledge Workers · Qualitative findings