Skip to content

Guide · 12 pages · 2026

AI in assessment item writing: a field guide

How we constrain, review and calibrate model-drafted assessment items so the reject rate falls below one in five: prompts, review rubric and pilot statistics.

Get the file
GuideAI in assessment item writing: a field guide

What is inside

  • A prompt structure that carries the standard, level and distractor policy.
  • The four failure modes reviewers look for first.
  • A calibration loop that feeds pilot statistics back into generation.

Get your copy

Tell us where to send it. The link arrives by email and works for 24 hours.

No sequences. A person may follow up once.