Deployment

Research with AI Agents: Find, Compare, and Cite Evidence

Create a traceable evidence memo by comparing sources, preserving citations, and exposing uncertainty.

Evidence stage · Module 3 of 15

Outcome for this lesson

Create a traceable evidence memo by comparing sources, preserving citations, and exposing uncertainty.

Bring forward: Bring the project and review boundary from Modules 1–2.

Produce: A one-page evidence memo with three claims, source links, confidence, and next verification actions.

0 of 15 core modules complete

Progress is saved on this browser. Create a free membership to sync across devices.

Recommended prerequisite: complete Module 2, “Vibe Coding Is Not Magic: Set the Human Review Boundary,” and bring its artifact. You may still use this page as a reference.

Evidence stage

Build the next result

Before interviewing people or building, create a traceable view of what is already known. AI can help locate and organize evidence, but the memo must preserve sources and uncertainty.

Your result for this module

A one-page evidence memo with three claims, source links, confidence, and next verification actions.

Continuous case study

Worked example: A traceable market claim

Weak note: “Freelancers often miss follow-ups.” Better note: “Four of five dated interview notes mention inconsistent follow-up; confidence is medium because the group is small; verify with a landing-page or concierge test.”

Do the work

  1. Write three decision questions

    Ask questions that could change the project, such as who experiences the problem, how they handle it now, and what failure costs them.

  2. Collect more than one source type

    Use primary documentation, public data, direct observations, and interviews where appropriate. Do not treat repeated summaries as independent evidence.

  3. Record each claim

    Keep the claim, direct link or note, retrieval date, method, confidence, disagreement, and next verification action together.

  4. Write the decision implication

    State whether the evidence supports interviewing, changing the audience, narrowing the promise, or stopping.

Copy-and-complete template

Create your module artifact

Decision question:

Claim:

Evidence and direct source:

Retrieved or observed on:

Confidence and why:

Contradicting evidence:

Next verification action:

Keep sensitive information, passwords, API keys, payment data, and private customer details out of course notes and AI prompts.

Quality gate

Check the result before the quiz

  • Every claim can be traced to evidence.
  • Dates and methods are recorded.
  • Uncertainty and disagreement remain visible.
  • The memo names a next test instead of pretending research is proof of demand.

How this moves the project forward

Module 4 combines this evidence memo with direct behavior to validate the problem and offer.

Evidence checkpoint and lesson summary

Confirm the artifact, then answer the key questions

The questions cover a core idea, a realistic decision, and evidence from the lesson. Incorrect answers identify concepts to review; they are not a complete measure of your experience or ability.

Artifact checkpoint

This is an honest self-check. Do not enter private data here; keep the artifact in your own approved workspace.

1. Two sources make different claims about the same market. What should the research workflow do?
2. Which research-agent output is most useful for a decision?
3. Which memo entry is ready for later validation work?

Confirm the artifact and answer all three questions correctly to unlock the recommended next step.

About the author

Son Kim

Son Kim writes from first-hand experience as a blind AI user. He tests vibe-coding tools, models, plugins, connected hardware, and everyday AI workflows to document what improves accessibility, independence, productivity, earning opportunities, problem-solving, and quality of life—and what still requires human judgment.

Read the full author background

Review the editorial, correction, advertising, and affiliate standards