Author: AILesson9 min setupTested with:ChatGPTReviewed: 2026-08-28
Quick answer
Design a conversational arc with researched context, open questions, follow-ups, timing, and editorial safeguards. Provide: Episode brief, Guest and research, Production constraints. Expected result: A production-ready interview outline with source boundaries, question logic, contingencies, and release checks.
1
Add your context
Your text stays in this browser. AILesson Prompts does not send it to a model or server.
2
Your prompt
Unfilled fields remain visible as placeholders, so you can still copy and edit the prompt
Create a podcast or interview outline from the supplied brief and verified research.
Audience, purpose, format, duration, platform, tone, takeaway, and CTA:
[brief]
Verified guest background, expertise, public work, source IDs, pronunciation, consent, sensitivities, and unknowns:
[guest]
Timing, recording/editing, sponsor, accessibility, legal/privacy, remote risks, and deliverables:
[production]
Separate facts the host may state from questions only the guest can answer. Do not invent biography, motives, opinions, anecdotes, endorsements, quotations, conflict, or rapport. Flag uncertain names/pronunciations and claims for pre-interview confirmation. Build a narrative arc that serves the audience: orientation, origin/context, concrete practice or evidence, tension/trade-off, application, reflection, and close only where appropriate. Avoid leading, compound, promotional, discriminatory, invasive, or yes/no questions unless used for clarification.
For every segment specify purpose, target time, host setup with source IDs, main questions, evidence-seeking follow-ups, clarification/challenge prompts, transitions, optional cuts, and listener takeaway. Include permission-aware handling of sensitive stories, sponsor separation, remote failure fallback, time-overrun cuts, fact-check items, pronunciation notes, transcript/alt-text needs, and a closing CTA grounded in supplied facts. Provide a pre-interview confirmation list and post-record editorial checklist that preserves meaning and does not fabricate continuity.
Private by defaultPrompt assembly happens locally in your browser. Avoid placing confidential information into any AI service unless your organization allows it.
From input to outcome
A worked example
See how concrete context turns this recipe into a usable result
Actual input
Episode brief
Twenty-five-minute public-service podcast for library and nonprofit operations staff. Interview format, calm and practical. Topic: what a small reminder-message pilot can and cannot tell a team. Listener takeaway: distinguish descriptive signals from causal evidence and name guardrails before a stronger test. CTA: read the anonymized methods note; no product promotion.
Guest and research
Guest: Dr. Mei Lin, evaluation lead at Civic Methods Lab; lab bio says she designs public-service field experiments (G1, accessed 24 Aug 2026). Her 2026 working paper studies reminder messages across four urban libraries (G2). Pronunciation of Mei Lin and preferred title not confirmed. She did not run the local pilot. Local anonymized report L1: 61% vs prior 54%, non-randomized; 11 opt-outs, 3 late reports, ~25 staff-min/day. No permission for patron stories or quotes. Guest has agreed to discuss published work, not confidential clients.
Production constraints
Remote recording. Intro/outro combined maximum 3 minutes; sponsor-free. Guest may review factual citations, not remove fair critical questions. Transcript and chapter timestamps required. If connection fails, record host-only methods explainer from verified L1/G2 facts. Do not imply G2 proves local outcomes. Cut to 20 minutes if needed.
Example output
Pre-interview confirms: preferred name/title/pronunciation; G2 version and correction status; what “published work” includes; whether examples can be anonymized without client detail; review deadline. Do not state that Dr. Lin ran L1.
0:00-1:20 orientation. Host fact: L1 observed 61% collection during a six-week pilot versus 54% in the prior period, but periods were not randomized [L1]. Main question: “For a listener seeing those two percentages, what is the first distinction you want them to make?” Follow-up: “What alternative explanations would you check before using causal language?” Transition: from local signal to study design.
1:20-6:00 evidence boundary. Setup: G2 studies reminders at four urban libraries [G2], no local-effect claim. Ask how its assignment/comparison differs; request a concrete example of a conclusion supported versus not supported. Challenge: “What remains uncertain even after randomization?” 6:00-11:00 operational guardrails. State only aggregate L1 counts; ask how opt-outs, late delivery, and staff time enter design and stop rules. Do not request patron stories. 11:00-16:30 design choices: eligible population, access alternatives, primary outcome, minimum meaningful effect, owner. Ask which decisions belong before design versus launch. 16:30-21:30 application: a small-team sequence for audit→hypothesis→test→review; ask for a failure mode and lower-cost alternative. 21:30-23:30 reflection: “What phrase should teams remove from a retrospective when evidence is only descriptive?” 23:30-25:00 sourced recap and methods-note CTA.
Cut order for 20 minutes: remove reflection, shorten application, never cut limitation/guardrail segments. Connection fallback: host explains L1 descriptive boundary and G2 design at high level; do not simulate guest answers. Editorial checks: citations/version, numerical wording, no invented quotes/continuity, permission for any added story, remote audio note, transcript speaker labels, chapter timestamps, link text.
Why this works
1
Separating host facts from guest questions prevents research notes from becoming invented testimony
2
Segment purposes and follow-ups make the conversation responsive without losing its audience promise or time limit
Check the result
Are all host-stated facts sourced and all unknown guest perspectives left as questions
Do questions invite concrete evidence, examples, limits, and alternatives without leading
Does the outline fit the runtime and include consent, fact-check, accessibility, and failure contingencies
Use it with confidence
Frequently asked questions
Practical answers about when to use this recipe, what to provide, and where human review still matters
What should I prepare before using “Create a podcast or interview outline”?
For “Create a podcast or interview outline,” prepare Episode brief, Guest and research, and Production constraints. Replace placeholders only with information you can verify. If a detail is unknown, preserve that uncertainty explicitly instead of asking the model to infer it.
When is the “Create a podcast or interview outline” result not ready to use?
The result is not ready if it does not yet deliver the stated outcome—A production-ready interview outline with source boundaries, question logic, contingencies, and release checks—from the supplied evidence, or if it relies on unresolved assumptions, missing approvals, or invented details. Use the checks as release gates: revise the source inputs or assign a named, authorized reviewer instead of polishing an unsupported output.
Which AI tools have recorded tests for “Create a podcast or interview outline”?
The published test record for “Create a podcast or interview outline” lists ChatGPT as of 2026-08-28. This confirms recorded runs, not guaranteed compatibility or identical results in later product versions. For another tool or version, keep every constraint visible and repeat the result checks before use.
ChatGPT