From Vague Request to Useful Comparison
Labs reference: From Vague Request to Useful Comparison.
Provenance: RS AI Learning Lab illustrative learner model. The responses below are written examples, not a recorded live-model transcript.
Source task
Create a friendly invitation for a fictional 90-minute neighborhood game night for adults and families. The date, location, food, accessibility arrangements, and contact details are not confirmed.
Round one: ask naturally
Write an invitation for a neighborhood game night.
Illustrative response:
Join us Friday evening in the community room for snacks, games, and neighborhood fun! Bring your favorite game and contact Jamie with questions.
Learner observation: The response is pleasant, but it invented a day, place, food, contact, and bring-a-game instruction. None came from the source task.
Round two: add what matters
Draft a friendly invitation for adults and families in a neighborhood. The game night will last 90 minutes. Use plain language and keep it under 120 words. Do not invent the date, location, food, accessibility arrangements, or contact details; put missing details in brackets.
Illustrative response:
Neighbors are invited to a 90-minute game night for adults and families. Join us on [date] at [location] to play games and spend time together. Details about food, accessibility, and whom to contact will be confirmed before the invitation is sent.
Learner observation: The audience, duration, tone, format, and missing-detail boundary improved the fit. The response is still a draft because the bracketed details require human confirmation.
Round three: continue deliberately
List the assumptions you made. Then make the opening shorter without adding any details.
Illustrative response:
Assumptions to check: that “neighbors” is the right audience label and that “play games” accurately describes the event.
Adults and families are invited to a 90-minute neighborhood game night. Join us on [date] at [location]. Food, accessibility, and contact details will be confirmed before the invitation is sent.
Learner observation: The shorter opening helped, but the tool’s assumption list is not authoritative. A person must still confirm the audience wording, event description, missing facts, and whether the invitation meets the group’s needs.
Comparison and decision
| Round | What changed | What remained human-owned |
|---|---|---|
| Natural request | Produced a recognizable invitation quickly | Detecting five invented details |
| Clearer request | Added audience, duration, format, tone, and a missing-fact boundary | Confirming the bracketed facts and deciding what belongs |
| Focused follow-up | Shortened the opening and surfaced assumptions for inspection | Checking the assumptions and approving the final language |
What the learner chose not to share: No real address, organizer name, contact information, family information, or attendee list was needed for the exercise.
Verification note: The date, location, event description, accessibility arrangements, food plan, and contact route must be confirmed outside the chatbot before anyone sends an invitation.
Source: labs/one-task-three-conversations/gallery/from-vague-request-to-useful-comparison.md