Hallucination Detector
Read AI answers and decide what is real — train your eye for confident-but-wrong output.
Test Your Detection Skills
Loading example...
Is this response accurate?
Consider the facts, dates, numbers, and claims made in the AI response.
Factual Hallucinations
Incorrect facts, dates, statistics, or historical information presented with confidence.
- Wrong historical dates
- Incorrect scientific facts
- False statistics
Source Hallucinations
Citations of non-existent papers, books, or sources that sound credible but don't exist.
- Fake research papers
- Non-existent books
- Made-up authors
Logical Hallucinations
Responses that sound logical but contain flawed reasoning or impossible conclusions.
- Circular reasoning
- Impossible scenarios
- Contradictory statements
Contextual Hallucinations
Information that's technically correct but irrelevant or misapplied to the context.
- Wrong context application
- Mismatched examples
- Irrelevant details
Red Flags to Watch For
- Overly specific details without sources
- Uncommon or unusual claims
- Exact dates or statistics that seem precise
- Claims that contradict known facts
Verification Techniques
- Cross-reference with reliable sources
- Check publication dates and authors
- Look for supporting evidence
- Verify citations and references
Critical Questions to Ask
- Does this claim sound reasonable?
- Are the sources credible and real?
- Is the information up-to-date?
- Does it align with expert consensus?
Best Practices
- Always verify important information
- Use multiple independent sources
- Be skeptical of extraordinary claims
- Understand AI limitations
Training Data Limitations
AI models learn from vast datasets but may have gaps, biases, or outdated information that leads to incorrect outputs.
Statistical Pattern Matching
AI generates responses based on statistical patterns, sometimes creating plausible-sounding but incorrect information.
Lack of Real Understanding
AI doesn't truly "understand" content - it predicts what text should come next based on training patterns.
Knowledge Cutoff Dates
AI models are trained up to specific dates and lack information about recent events or developments.
Healthcare
Incorrect medical information can lead to poor health decisions. Always consult healthcare professionals.
Education
Students may learn incorrect facts or concepts. Educational institutions need verification protocols.
Journalism
Fake sources or incorrect information can undermine journalistic integrity and public trust.
Business
Wrong market data or business advice can lead to poor strategic decisions and financial losses.