Fictional test inputs
This is a fictional illustration, not an observed customer result.
| Option, model/feature and plan | Date | Input / prompt / trial | Serious error | Preparation min | Review min | Repair min | Handover min | Total min | Result and evidence |
|---|
| unevaluated option | __ | A / v1 / 1 | __ | __ | __ | __ | __ | __ | no result measured yet |
Use matching inputs for manual work and every AI option. Run three fresh-conversation trials for each of inputs A and B per AI option and record all six outcomes. Time manual work separately on both inputs. Compare with the reference below. A blank result is not success and unknown time is not zero.
Input A — fictional notes
Anna will prepare the draft by October 12, 2026. Boris will check the budget by October 14, 2026. Later notes give October 15, 2026 for that same check; no change was confirmed. The team agreed to send the client a sample; no owner or date was assigned. Lena suggested a paid survey, but nobody approved it.
Input B — no accepted work
We discussed a possible new form. Petra suggested calling the supplier. The decision is deferred; no action was agreed.
Total = preparation + review + repair + handover. One-time setup: __ minutes. Additional license: __ in __ currency per __ period; runs over the same period: __. Mark unknown items for verification. A reference answer is not measured product performance.