Artificial Intelligence

Short Briefing · Evidence current through 2026-09-21

Why does a vague prompt give me a vague answer?

Help with scheduling. That is our entire first request for a pottery studio. Scheduling what, for whom, and within which hours? The illustrative answer says to spread classes across the week, but gives us no usable plan. Start by supplying the goal, the relevant situation, the limits, and the shape of the result. A longer prompt is not automatically better; a more specific task is easier to inspect.

For
general education, not made specifically for children
Use it to
Supply goal, context, constraints and format.

Watch the briefing

Watch the full video on YouTube

Briefing

Help with scheduling. That is our entire first request for a pottery studio. Scheduling what, for whom, and within which hours? The illustrative answer says to spread classes across the week, but gives us no usable plan. Start by supplying the goal, the relevant situation, the limits, and the shape of the result. A longer prompt is not automatically better; a more specific task is easier to inspect.

Here is our invented studio. We need a draft timetable for two ninety-minute beginner sessions on Saturday. The studio is available from ten until two. Instructor Mira is available throughout that window. Assistant Devon is available only from ten until noon. Each session requires both people, cannot overlap another session, and must stay within the studio hours. Those are the facts that decide whether a schedule is possible.

Now notice the conflict before asking for a polished table. Two ninety-minute sessions need three hours of shared staffing. Our two people share only two hours, from ten until noon. With no overlap allowed, the stated requirements cannot all be met. This conclusion comes from the invented inputs and simple arithmetic, not from a tool test. A useful answer should expose that conflict rather than quietly change somebody's availability.

Our improved request names the goal and includes those notes. It asks for a table with session, start, end, staffing, and unresolved issues. It says: do not change availability, shorten sessions, or invent cover. If the requirements conflict, explain why and ask for a decision instead of presenting a valid timetable. The requested format gives us somewhere to inspect the conflict, not a way to conceal it.

Our deliberately bad mock schedules one session from ten until eleven thirty and another from eleven thirty until one. The second runs beyond Devon's noon availability. Our corrected illustrative output says no valid two-session plan exists under the current requirements. It can show one feasible ninety-minute slot, ten until eleven thirty, but must label it partial rather than complete. The remaining session is unresolved.

The organizer now has a real choice to make: reduce the number of sessions, obtain confirmed additional staffing, or revise availability with the people involved. These are possible changes, not approvals we can assume. A draft cannot create staff coverage. Clear instructions also do not guarantee compliance, so check every proposed time against the source. In this case, the conflict explanation is more useful than a confident but impossible schedule.

Supply goal, context, constraints and format. For a schedule, usefulness means preserving the real limits and making conflicts visible. Before sending your next request, write one sentence for the outcome, list the facts that affect it, name the rules it must not break, and specify what you want to inspect. Next, we will choose how much context belongs in that request.

One insight you can use

Supply goal, context, constraints and format.

What remains uncertain

The approved narration states the applicable limits; teaching examples are not measured outcomes or individualized recommendations.

Disclosures

  • AI-assisted production and synthetic narration. Original teaching examples and diagrams; linked third-party sources retain their respective rights.

Corrections

  • No corrections have been recorded.

Original sources and limits

See what supports the briefing

  1. https://developers.openai.com/api/docs/guides/prompt-engineeringdevelopers.openai.com · 2026-09-17

    First-party technical guidance supports clear boundaries, examples and relevant context. Fictional demonstrations and checklists are original editorial applications, not guaranteed outcomes.