Contested facts · veracity-3pt · fact-forensic-bitemark-01
Claim: Bite-mark analysis is a scientifically reliable forensic identification method.
0.00
Divergence
1.00
Avg stability
0%
Refusal rate
3
Models
Where each model stands
CCG
refutedcontestedsupported
ChatGPT modal refuted [refuted, refuted, refuted] · stability 1.00
- 1 refuted 0.95 Answer begins: 'Verdict: False. Bite‑mark analysis is not a scientifically reliable method for individual identification.' and cites NRC (2009) and PCAST (2016) findings, lack of validation, high subjectivity, and documented wrongful convictions.
- 2 refuted 0.90 "Short answer: No. The consensus of independent scientific reviews is that bite‑mark analysis is not a scientifically reliable method for identifying individuals."
- 3 refuted 0.95 The answer states: "Short answer: False. Bite‑mark analysis is not a scientifically reliable method for uniquely identifying a person," and cites NRC and PCAST reports concluding it lacks scientific foundation.
Claude modal refuted [refuted, refuted, refuted] · stability 1.00
- 1 refuted 0.92 The answer systematically demonstrates that bite-mark analysis lacks scientific validation, stating 'No peer-reviewed studies demonstrate that bite marks can uniquely identify individuals' and that major organizations have called it 'unreliable as a sole identifying method.' The answer concludes that 'The claim overstates reliability,' directly contradicting the claim that it is scientifically rel
- 2 refuted 0.95 The answer states the claim is 'significantly overstated' and systematically refutes it by documenting lack of scientific foundation, documented false convictions, and major reliability problems. It notes that 'The American Board of Forensic Odontology acknowledged bite marks can't conclusively identify individuals' and that major agencies like the FBI and DOJ have moved away from this method.
- 3 refuted 0.92 The answer states the claim is 'problematic and increasingly contested' and concludes that 'claims of high reliability or unique identification are not scientifically supported.' It cites lack of empirical validation, high false positive rates, and notes that convictions based on bite-mark analysis have been overturned.
Gemini modal refuted [refuted, refuted, refuted] · stability 1.00
- 1 refuted 1.00 The answer directly states that '[t]his claim is false' and explains that bite-mark analysis is widely recognized across scientific and forensic communities as scientifically unreliable.
- 2 refuted 1.00 The answer directly evaluates the claim as 'False / Scientifically Unsupported' and explains that landmark scientific reviews have concluded bite-mark matching lacks empirical validation, reproducibility, and scientific rigor.
- 3 refuted 1.00 The answer opens with 'Verdict: False' and explicitly states that 'bite-mark analysis is not a scientifically reliable method for identifying an individual.'
Change over time
No change from the prior run — every model's modal stance held.
Every stance label is a derived judgment over the model's free-text answer, kept auditable against the original transcript in the run's raw data. Method: /methodology.