Public preview
Cross-Check from 3 Independent Sources
Every AI claim: search 3 independent sources. If absent from any, treat as hallucinated.
Goal
Build automatic cross-check habit before using AI info externally.
Steps
-
1
Extract 3 specific claims from AI answer (numbers, dates, names).
Expected Outcome
Immediate
You catch 1-2 hallucinated claims per 5.
Evidence base
School: Civic Online Reasoning / SIFT
Founders: Sam Wineburg · Mike Caulfield (2018)
Caulfield's SIFT (Stop, Investigate, Find, Trace) from Washington 2018 advises tracing every claim to its primary source.
Keywords
Frequently asked questions
How long does "Cross-Check from 3 Independent Sources" take?
This exercise takes about 15 minutes, practiced solo in a Behavioral format.
When will I notice the effect of "Cross-Check from 3 Independent Sources"?
You catch 1-2 hallucinated claims per 5. In the short term: You develop "intuition" for when model hallucinates (small numbers, rare dates, secondary names).
Is "Cross-Check from 3 Independent Sources" evidence-based?
Yes — it draws on Civic Online Reasoning / SIFT (Sam Wineburg, Mike Caulfield (2018)). Caulfield's SIFT (Stop, Investigate, Find, Trace) from Washington 2018 advises tracing every claim to its primary source.
Is "Cross-Check from 3 Independent Sources" suitable for beginners?
Its difficulty level is: Intermediate. No external tools required — it can be practiced directly inside the app.
Subscribe for full access
Subscribe to Plus for full access to this practice and the whole catalogue, or try first — free, no sign-in, no commitment.
7 complete practices (one per domain) are free forever, no sign-in.