Description
Write questions sharp enough to expose what a source actually knows.
Challenge
- Challenge 01
Take a claim and write three interrogating questions that a weak source would fail: one for evidence, one for source, one for what it leaves out. Avoid yes/no questions.
- Challenge 02
A source says "the data shows a trend." Write two questions: one that probes whether the trend is real or an artifact of how the data was collected, one that asks what comparison would make the trend meaningful or meaningless.
- Challenge 03
Your source claims an AI system is "biased." Write one question that forces them to name the specific group harmed, one that forces them to name the specific decision the bias affects, and one that asks how they measured it.
- Challenge 04
A source claims a feature is "user-friendly" or "easy to use." Write questions that expose: what task was tested, what comparison group used it (experts vs. novices?), and what "easy" actually measured.
- Challenge 05
A headline claims AI will "replace" workers in a field. Write three questions: one about timescale (how long?), one about scope (which tasks, not which people?), and one that probes what the source actually measured versus predicted.
- Challenge 06
A source offers a statistic like "90% of teachers support this approach." Write questions that expose the survey method, the exact wording of the question, and what options were presented.
- Challenge 07
A source cites a case—"Company X did Y and profits surged." Write one question that isolates causation, one that asks what else changed at the same time, and one that probes whether the result generalizes.
- Challenge 08
A source claims an AI model is "more accurate" than another. Write questions that expose: accurate at what task, on what data, compared to what baseline, and measured how.