Basic draft verification. No coding required. Use the shared foundation if AI tools are unfamiliar.
A browser and text editor. Live source discovery needs browsing; all supplied exercises also work offline.
Start with the idea
Turn a broad topic into a question you can answer.
Start with the decision your reader needs to make. State which place, people and period the question covers. List the missing evidence before asking for a recommendation.
Research brief: A short plan stating the question, audience, scope and evidence needed.
See the example and check your understanding
A made-up teaching example
A library has eighteen replies favoring evening workshops. The total respondent count and available staffing are unknown. Ask whether one branch can run a pilot, not whether the whole city wants evening events.
Ask yourself
Do eighteen favorable replies prove citywide demand?
Show the explanation
No. Check how replies were collected and who they represent. Staffing and demand are separate questions.
Start with these checks
- Name the decision and scope.
- List evidence gaps.
- Do not generalize beyond the sample.
Use the different practice case below and check every requirement in its full instructions.
Full learning objectives
- Specify audience, decision, scope and date.
- Distinguish discovery from verification.
- Choose a tool configuration that can access the needed evidence.
Captions appear here during playback.
Turn a broad question into a research brief
Turn a broad topic into a question you can answer.
Video playback is unavailable. Read the descriptive transcript below, go directly to practice, or open the video file.
Source-based teaching example · Fictional data · Reviewed 2026-10-03. Microsoft Mark · locally synthesized. On a small screen, use fullscreen for diagram details or read the transcript.
Read the descriptive transcript
0:00 · What you will learn
Turn a broad topic into a question you can answer.
What appears on screen
Turn a broad topic into a question you can answer.
0:05 · The idea
Start with the decision your reader needs to make. State which place, people and period the question covers. List the missing evidence before asking for a recommendation.
What appears on screen
Start with the decision your reader needs to make. State which place, people and period the question covers. List the missing evidence before asking for a recommendation.
0:19 · One useful term
Research brief. A short plan stating the question, audience, scope and evidence needed.
What appears on screen
A short plan stating the question, audience, scope and evidence needed.
0:29 · A worked example
A library has eighteen replies favoring evening workshops. The total respondent count and available staffing are unknown. Ask whether one branch can run a pilot, not whether the whole city wants evening events.
What appears on screen
A library has eighteen replies favoring evening workshops. The total respondent count and available staffing are unknown. Ask whether one branch can run a pilot, not whether the whole city wants evening events.
0:45 · Your turn to think
Do eighteen favorable replies prove citywide demand? Pause and decide before continuing.
What appears on screen
Do eighteen favorable replies prove citywide demand? Pause the video if you need more time.
1:02 · Why this answer works
No. Check how replies were collected and who they represent. Staffing and demand are separate questions.
What appears on screen
No. Check how replies were collected and who they represent. Staffing and demand are separate questions.
1:12 · Check 1
Name the decision and scope.
What appears on screen
Name the decision and scope.
1:17 · Check 2
List evidence gaps.
What appears on screen
List evidence gaps.
1:21 · Check 3
Do not generalize beyond the sample.
What appears on screen
Do not generalize beyond the sample.
1:27 · Now try a different case
Open the lesson and try its different practice case. Use every check in the full task. Watching alone does not mark the lesson complete.
What appears on screen
Open the lesson Try the new case Check and revise The full task and its requirements are in the linked lesson.
Watching explains the example. Complete the lesson practice to check what you can do. Explore all visual lessons →
The method
Begin with the decision your reader must make. “Research AI” has no stopping rule. A useful brief names the question, the population or process, the period, the output, and which facts would change the recommendation. Separate what is known from assumptions before asking for a draft.
Ask an assistant to identify missing information and propose search terms before it writes an answer. Browsing is a capability of the application or harness, not an automatic property of a powerful model. If it cannot browse, provide documents and restrict the answer to them. A supplied-document exercise practices synthesis; it does not establish current external facts.
Choose your model, harness and offline route
Text-only assistant
Use supplied text. Ask for a plan or draft; do not claim browsing, code execution or file access that was unavailable.
Tool-enabled application or harness
Record which tools actually run, allowed files and actions, approval boundaries and observable results. Tools are optional unless the task needs them.
Local or cloud deployment
Inspect inference, retrieval, connectors, logs and storage separately. Use only authorized environments for real data; training fixtures are fictional.
Higher-capability or frontier model
Evaluate the exact task and configuration. Capability does not supply missing evidence or grant permission. No provider ranking or guarantee is implied.
Offline alternative
Use the supplied inputs and inspect the worked example manually. Builder exercises use local deterministic Python. No paid account or live model call is required.
Record the exact product/model label when available, tools actually used, inputs, permitted actions, observed output and checks. Live accounts or APIs may have external costs; they are optional and are not used by this page.
Read the general workflow guide →Worked example
Supplied input
Fictional brief A: A library is considering evening workshops. Audience: program manager. Decision due 15 November. Scope: this branch only. Known: 18 survey replies favor evenings; respondent count and venue staffing are unknown.
A checked result
Question: Is there enough evidence to pilot one evening workshop at this branch? Request survey denominator, response method, staffing and a pilot success criterion. Deliver a one-page options brief; do not infer citywide demand from 18 replies.
Why this works
The question narrows the decision and identifies missing evidence. Favorable replies alone do not establish representativeness or feasibility.
Try a different case
Independent brief B: A fictional museum is considering Saturday tours. Nine visitors requested tours; 120 visited during the sampled week. Staff availability and the collection method are unknown. Write a research brief, not a recommendation.
Download this task and checklist →Interactive practice / Fictional fixture
Build your research brief
Use the supplied independent case above. Check the bounded decisions, then review your written artifact against the full acceptance criteria.
Everything runs in this browser. No AI model is called and no work is uploaded. Bounded checks cover specific facts and rules. Your writing is not graded: you review it yourself. Practice never marks a lesson complete automatically.
Write an answerable question for this museum, not a recommendation.
Explain which missing facts could change the decision and when you would stop.
Try the task, run the checks, then revise. Checks are teaching feedback, not an instructor grade.
Observed browser simulation
This trace comes from the bounded local checker. It is not a Python run, model result or production tool execution.
Self-review
Locked. Write every written part, then run the checks. A few words is not an attempt.
This is self-review, not grading. Read our answers, then tick what your own work does. Ticks are optional. They stay on this page and are not saved or sent anywhere. A different answer can be just as good if it meets the lesson criteria.
Your decision question, audience and scope
A strong answer
Question: Is there enough evidence to justify a trial of one Saturday tour at this museum? Audience: the museum manager who decides on programs. Scope: this museum, Saturday tours only, based on one sampled week. Known: nine visitors asked for tours; 120 visited that week. Not known: how the requests were collected, and whether staff can cover a Saturday. Deliverable: a one-page brief with options. No recommendation yet.
A flawed answer
7.5% of museum visitors want Saturday tours, so the museum should start running them. Audience: everyone at the museum. Scope: visitor demand.
What is wrong with it: It turns nine requests in one week into a rate for all visitors, and nobody knows how those requests were collected. It also recommends before it asks. The task was a brief, and staffing is still unknown.
Evidence gaps and stopping rule
A strong answer
Gap 1: how the nine requests were collected. If staff only wrote down people who asked at the desk, the number says little about demand. Gap 2: whether staff can cover a Saturday. If they cannot, demand does not matter yet. Gap 3: whether the sampled week was typical. Stop when I have the collection method and a staffing answer. If either cannot be had in two weeks, report that as the finding and propose a small trial with a headcount.
A flawed answer
We need more data on visitors. I will keep researching until I am confident that tours will be popular.
What is wrong with it: “More data” names no fact that could change the decision. “Until I am confident” is not a stopping rule: it has no end, and it is looking for one answer.
Keep or restore your work
Drafts are not saved automatically. Save only fictional practice data. A browser draft may be lost if storage is cleared; download a copy to keep it.
Your steps
- Name the decision, audience, scope and output.
- List three evidence gaps and how each could change the decision.
- Write a bounded AI request with explicit unknown handling and a stopping rule.
Check your work
- Do not treat 9/120 as a representative preference estimate.
- Separate demand, staffing and feasibility questions.
- Specify how sources and unsupported claims will be recorded.
After your attempt: review notes and a common mistake
A defensible brief asks how visitors were sampled and whether staffing supports a trial. “7.5% of all museum visitors want tours” overgeneralizes one sampled week and an unknown collection method.
Different wording or a different justified decision can pass. Check your work against every requirement; these notes are not an automated grade.
Explain your approach
Which missing fact would change your next research step most, and why?
Keep your artifact, one correction or decision, its evidence and one remaining limitation. Nothing on this page sends or uploads your work.
Check your understanding
Choose an answer to see feedback. This is a knowledge check, not instructor assessment.
Review your independent practice
Complete the supplied task before recording progress. Use the acceptance criteria and reveal the review notes after your attempt. Keep your artifact yourself; nothing is uploaded.
Not yet self-reviewed. Completion stays in this browser and does not issue a badge.
Optional deeper study
Explore the related general-course lesson → This opens the existing canonical lesson; it is optional depth, not another copy of this focused task.
Primary references
Revision 2026-10-03. Fictional examples, not observed provider results. Knowledge checks and completion records are self-study aids.