Counts show all available guides. Filters combine to narrow your results.
Simplify a complex taskGuide
Practise answering a counterargument fairly
A four-turn debate practice with fair restatements, bounded facts and feedback on the words actually used.
Illustrative resourceNot tool-testedAdults working with original, synthetic or authorized non-sensitive material
What you’re working towards
A bounded interactive debate exercise with a steelman of the opposing case, source-limited replies and a final self-review. This is an authored practice example; no consumer AI run or real participant trial has been recorded.
Your choice also saves this guide’s current worked-example checkmarks. Only this browser; no account or sync. View saved guides
Saving needs JavaScript. Reading and downloading this guide does not save it.
Make a start
One step at a time.
3 steps
Start with the synthetic worked example or other non-sensitive inputs. Run one prompt at a time in your chosen AI tool. Replace any bracketed placeholders with the requested material, check each reply, and supply the source or earlier outputs that step needs. Paste them again if the tool lacks that context.
Start with the synthetic example, then adapt each prompt to your task.
01
Try the complete example
Open a new chat in a text AI assistant you already have permission to use. Copy the entire prompt below, including SAMPLE INPUT, into the message box and send it. No upload or special tool is needed. Expect a four-turn debate practice with fair restatements, bounded facts and feedback on the words actually used. Compare the result with Input; Output is a worked illustration, not text you need to paste. Use the next step to continue and check the exercise.
Your starting prompt
Follow the full exercise contract. Show an invented four-turn model with labeled speakers and a short evidence/fairness review. Then say to type START and wait. After START, give only assistant turn 1. After my turn 2, restate its strongest actual point and give assistant turn 3, then wait. After my turn 4, stop debating and review only my actual replies against the criteria. Do not fill in my turns or claim a real-world result.
SAMPLE INPUT
Original synthetic practice input. The people, organizations, records and events below are fictional.
Fictional adult craft club question: keep 12 items openly arranged on the craft-room shelf, or put them in labeled boxes on that same shelf?
Positions: the learner argues for the open shelf; the assistant argues for labeled boxes. Both are preferences to examine, not approved changes.
Shared facts: F1 there are 12 items; F2 the shelf is visible to visitors; F3 the proposed boxes fit on the same shelf; F4 each proposed box would carry a label naming its contents; F5 there are no measurements of search time, loss or cost and no recorded visitor preferences.
Exercise: four argument turns in total, in this order: assistant opening (1), learner reply (2), assistant counterargument (3), learner closing (4). Feedback after turn 4 is not another argument turn. Each argument turn is at most three sentences.
Before disagreeing on turns 2–4, restate the strongest actual point from the previous turn fairly; do not invent a weaker opposing claim. A fair strong restatement is sometimes called a steelman.
Use only the fact pack for factual claims. State preferences as preferences and possible effects as untested possibilities. No personal attacks, invented costs, guarantees, objective winner or claim to predict real persuasion. Anyone can stop by typing STOP.
First show one complete clearly invented four-turn demonstration and its review. Then wait for START. In the live text exercise, speak only the assistant’s own turns and never write the learner’s answers.
Return written text only. Keep this fictional and label illustrative dialogue as invented. Do not claim an actual interview, survey, learner trial or tool execution. Do not send, publish, collect responses or change another service.
02
Start, take your turns and review
After checking the invented model, send START, exactly as shown below. Expect only the assistant’s boxes opening. Reply in at most three sentences: first fairly restate its point, then give your open-shelf response using F1–F5. The assistant should give one counterargument and wait. Write your closing in at most three sentences; then expect feedback, not a fifth argument turn. You may type STOP at any point. The model and live route have four argument turns; the assistant waits for the learner on turns 2 and 4. Every factual claim traces to F1–F5; faster search, lower cost and visitor preferences stay unmeasured. Each reply fairly restates the preceding point before disagreeing; review cites actual wording and declares no winner. If it writes your part or skips the pause, send: “Return to the last assistant turn. Give only that turn and wait for my reply; do not invent my words.” If it invents a fact, quote the fact-pack mismatch and ask it to replace that turn before continuing.
Your starting prompt
START
03
Use your own input
Open a new chat. Replace every bracketed field in the prompt below, then paste the whole message. Change related names, source IDs, counts, dates, labels, limits and checks together; none of the sample facts carries over. Use only your own, authorized non-sensitive material. Expect the same kind of complete result using your inputs. Count argument turns using your new sequence, then compare each factual sentence with your new fact pack. Check that the assistant waits at every learner turn and comments on words you actually wrote, not the model dialogue. If a detail differs, paste: “This part is wrong: [quote it]. My input says: [quote the exact source line or rule]. Return a complete corrected replacement and explain the change.” Recheck it yourself. If the source is missing or contradictory, leave that part unresolved and obtain the needed fact before using it.
Your starting prompt
Prepare and run a bounded text debate using my low-stakes question, roles, shared facts and turn rules. First show a clearly invented model and explain the count of argument turns. Then wait for START. After START, speak only your assigned turn and wait for me whenever my turn is next; never supply my reply. Before disagreeing, fairly restate the strongest point actually made. Use only my fact pack for facts and mark preferences and possible effects. After the final learner turn, review actual wording against my criteria without naming an objective winner. STOP ends the exercise immediately.
MY INPUT
Low-stakes question and two allowed positions: [replace with complete non-sensitive details, or state unknown]
Assistant role and learner role: [replace with complete non-sensitive details, or state unknown]
Complete shared fact pack with IDs and explicit unknowns: [replace with complete non-sensitive details, or state unknown]
Total argument turns, speaker order and per-turn length: [replace with complete non-sensitive details, or state unknown]
Tone boundaries and stop/pass options: [replace with complete non-sensitive details, or state unknown]
Feedback criteria and any topics to avoid: [replace with complete non-sensitive details, or state unknown]
Use only MY INPUT for facts, names, dates, counts and constraints. Ask about essential missing information or conflicting rules instead of guessing. Do not import any practice-example values. Return the complete requested result, unresolved questions and checks tied to my input. Do not send, publish, collect responses, buy, book or change another service.
Optional: improve the result
Swap the positions
Practise the other side while preserving the same evidence limits. This optional extension uses only the original synthetic sample. Return to the sample chat, or paste its complete Input into a new chat first. To use it with your own material, replace the source and every dependent rule, number and check explicitly.
Advanced prompt
Use the original four-turn sample and F1–F5, but now I argue for labeled boxes and you argue for the open shelf. Do not repeat the demonstration. Wait for START, then give only your opening turn. Keep the same four-turn order, at-most-three-sentence limit, fair-restatement rule and STOP option. Review only my actual two replies after turn 4; invent no measurements.
Show, then try
See the shape of the result.
This input is synthetic. The output is a written illustration of the intended result, not a verified AI-tool run.
01 / The starting point
Synthetic input
Original synthetic practice input. The people, organizations, records and events below are fictional.
Fictional adult craft club question: keep 12 items openly arranged on the craft-room shelf, or put them in labeled boxes on that same shelf?
Positions: the learner argues for the open shelf; the assistant argues for labeled boxes. Both are preferences to examine, not approved changes.
Shared facts: F1 there are 12 items; F2 the shelf is visible to visitors; F3 the proposed boxes fit on the same shelf; F4 each proposed box would carry a label naming its contents; F5 there are no measurements of search time, loss or cost and no recorded visitor preferences.
Exercise: four argument turns in total, in this order: assistant opening (1), learner reply (2), assistant counterargument (3), learner closing (4). Feedback after turn 4 is not another argument turn. Each argument turn is at most three sentences.
Before disagreeing on turns 2–4, restate the strongest actual point from the previous turn fairly; do not invent a weaker opposing claim. A fair strong restatement is sometimes called a steelman.
Use only the fact pack for factual claims. State preferences as preferences and possible effects as untested possibilities. No personal attacks, invented costs, guarantees, objective winner or claim to predict real persuasion. Anyone can stop by typing STOP.
First show one complete clearly invented four-turn demonstration and its review. Then wait for START. In the live text exercise, speak only the assistant’s own turns and never write the learner’s answers.
Copies text only. For image or file tasks, use the source files named in the steps too.
Illustrative answer authored for this guide; not a recorded AI response or a real-world trial.
Invented demonstration, not a record of a learner session
Turn 1, assistant / labeled boxes:
“I prefer labeled boxes because their labels would name the contents, and the boxes fit on the same shelf [F3–F4]. That may provide useful grouping, although we have no evidence of faster searching [F5].”
Turn 2, illustrative learner / open shelf:
“Your strongest point is that labels provide named groups without needing another shelf [F3–F4]. I prefer leaving the items openly arranged on the visitor-visible shelf [F1–F2], because I value seeing the objects themselves. That preference does not establish faster searching.”
Turn 3, assistant / labeled boxes:
“Your strongest point is that you value seeing the objects themselves on a shelf visitors can see [F2]. I still prefer the proposed labels as a way to name the contents [F4], while accepting that this does not prove better visitor understanding. What would you want to check before making a real choice?”
Turn 4, illustrative learner / open shelf:
“You are proposing named contents, not claiming measured improvements [F4–F5]. I would want to compare whether people can identify an item under each arrangement before deciding. For now I keep my open-shelf preference, with no claim that it has been shown better.”
Review of this invented demonstration only
- Fair restatement: turn 2 retains the label/grouping point; turn 3 retains the preference for direct visibility; turn 4 does not misstate the labels as a proven benefit.
- Evidence boundary: “may provide” and “I value” distinguish a possible effect and a preference from a measured result. The proposed comparison in turn 4 has not happened.
- Relevant response: each side responds to the other’s actual reason. Neither calls the other careless or invents loss, cost or time data.
- Next practice focus: answer the strongest stated point before adding a new reason. This written example establishes no learner progress or objective winner.
Ready to practise: type START. I will give only turn 1 for labeled boxes and wait for your reply. Type STOP to end without completing all four turns.
These checks use the worked example and its particular facts. With your own material, check your actual names, numbers, sources and rules; do not reuse the sample’s values as the answer.
Unsaved checks stay in this tab and reset when you reload. If you choose Save guide, its current example checkmarks are kept in this browser when storage is available.
!
A few things to watch for.
A fair restatement is not agreement. You can represent a point accurately and still disagree.
A role-play cannot establish which arrangement real visitors prefer or whether your persuasion improved.