Researchers Discover Security Vulnerability in Forensic DNA Analysis Technology

Researchers from the University of New Haven and collaborating forensic specialists have identified a security vulnerability in technology widely used to analyze DNA evidence, raising concerns about the integrity of digital forensic records maintained by crime laboratories.

The vulnerability, known as DNAForge and identified as CVE-2026-17583, affects certain digital files generated by Thermo Fisher Scientific's Applied Biosystems human identification technology. Researchers demonstrated that ".fsa" and ".hid" files containing DNA analysis data could be modified without triggering reliable indications of tampering.

The research involved University of New Haven forensic science professor Laura Gaydosh Combs and Nathan Adams of Forensic Bioinformatics, alongside cybersecurity specialists. Their investigation revealed that the weakness could potentially affect digital DNA records generated by laboratory instruments dating back to the 1990s.

Researchers also demonstrated how generative artificial intelligence could simplify the manipulation process. Using Anthropic's Claude, they reportedly reduced the time required to modify an evidence file from approximately 45 minutes to three minutes.

The findings have significant implications for forensic laboratories because DNA analysis frequently contributes to criminal investigations, suspect identification, and courtroom proceedings. Manipulated electronic records could potentially misrepresent genetic profiles, although the vulnerability does not alter the original biological samples.

The researchers reported the issue in May 2026. Thermo Fisher Scientific subsequently issued a security bulletin in July and released software updates incorporating digital signatures to strengthen file integrity verification.

The company stated that it was unaware of any confirmed instances in which the vulnerability had been exploited to affect actual forensic casework.

However, researchers have cautioned that identifying historical manipulation remains challenging, particularly for files generated before the security updates. Existing records may require additional verification using preserved originals, laboratory documentation, and established chain-of-custody procedures.

The discovery highlights the growing intersection between cybersecurity and forensic biology, emphasizing that scientific evidence reliability depends not only on laboratory testing methods but also on the security of the digital systems used to process and preserve results.