Public preview
Hallucination Check: 3 Questions Before Accepting
Before accepting any AI answer, ask: source? dates correct? names real?
Goal
Prevent reliance on hallucinated info that comes out confidently but is wrong.
Steps
-
1
After any answer with dates/names/numbers/quotes: stop.
Expected Outcome
Immediate
You avoid likely professional embarrassment from 30% of AI answers.
Evidence base
School: AI Safety / Critical Thinking
Founders: Ji et al. · Anthropic Research (2023)
Ji et al. 2023 "Survey of Hallucination in NLG" shows LLMs hallucinate in 15-30% of detailed answers.
Keywords
Frequently asked questions
How long does "Hallucination Check: 3 Questions Before Accepting" take?
This exercise takes about 10 minutes, practiced solo in a Behavioral format.
When will I notice the effect of "Hallucination Check: 3 Questions Before Accepting"?
You avoid likely professional embarrassment from 30% of AI answers. In the short term: You master distinguishing "linguistic confidence" from "factual accuracy".
Is "Hallucination Check: 3 Questions Before Accepting" evidence-based?
Yes — it draws on AI Safety / Critical Thinking (Ji et al., Anthropic Research (2023)). Ji et al. 2023 "Survey of Hallucination in NLG" shows LLMs hallucinate in 15-30% of detailed answers.
Is "Hallucination Check: 3 Questions Before Accepting" suitable for beginners?
Its difficulty level is: Beginner. 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.