We've Been Using The Wrong Science In Court For 50 years

Forensics can be valuable but only if methods are continually validated because many traditional practices were adopted without rigorous testing, so ongoing empirical reassessment and guardrails are required to ensure techniques truly connect evidence to individuals.

Whole genome data can introduce bias because revealing phenotype and ancestry cues like hair color or ethnicity lets analysts form expectations that influence downstream interpretation and risk discriminatory conclusions.

DNA mixtures are hard to interpret because overlapping STR signals from multiple contributors create ambiguous peaks, and proficiency studies show labs often assign those peaks incorrectly.

As DNA detection becomes more sensitive, tiny amounts transferred indirectly can incriminate innocent people because secondary transfer on gloves or equipment can be detected and misattributed to someone absent from the scene.

When examiners see prior opinions or case details, conformity pressure biases them toward confirming earlier findings because knowledge of verdicts and stakes raises agreement and inflates perceived certainty.

Fingerprint database searching depends on subjective minutiae selection because examiners vary in how many features they mark, and only prints above a threshold are searchable, so that variability changes whether a match is even attempted.

Bloodstain patterns are ambiguous because the same-looking spatter can come from different actions and changing blood properties like viscosity alter droplet behavior, so analysts can legitimately disagree about origin.

Simple trigonometric bloodstain analysis can point to the wrong origin because treating stain trajectories as straight lines ignores gravity and aerodynamic drag, which shift true paths.