Axon Enterprise's Draft One, a generative AI product designed to write police report narratives from body-worn camera audio, has been scrutinized by the Electronic Frontier Foundation (EFF). The investigation reveals that the product may hinder efforts towards auditing, transparency, and accountability.
The EFF's findings are based on public records from various police agencies using Draft One, along with Axon's user manuals and other materials. The investigation highlights that the product offers limited oversight features. Consequently, when a police report contains biased language or inaccuracies, it is unclear whether these issues originate from the officer or the AI. This lack of clarity complicates assessments of how the system impacts justice outcomes over time.
“Police should not be using AI to write police reports,” stated Matthew Guariglia, EFF Senior Policy Analyst. He expressed concerns about unanswered questions regarding AI translation of audio situations and whether edits will be made by officers. He also questioned if the public could discern between human-written and computer-generated content in reports.
Axon’s technology bundles, including Tasers and body-worn cameras, are utilized by numerous police agencies. The company leverages these relationships to promote Draft One extensively. More cities are expected to adopt this AI in upcoming years.
The EFF discovered that Draft One does not save generated drafts or any edited versions. Officers copy the AI draft text into the police report, after which the draft disappears upon closing the window. This leaves no trace for judges, defense attorneys, or the public to determine which parts were written by AI versus those written by officers.
Although Axon has advertised its “audit log” function as a transparency measure, EFF found it provided minimal insight into technology usage. Nonetheless, EFF released a guide on records obtainable under public records laws.
“As AI technology proliferates in policing, it’s crucial that journalists, researchers, and advocates try to get these records to not only identify poor police practices but also to highlight structural gaps in accountability,” said Dave Maass, EFF Director of Investigations.
For more information on EFF’s Axon Draft One investigation: https://www.eff.org/deeplinks/2025/07/axons-draft-one-designed-defy-transparency
For EFF’s guide on requesting public records about Axon Draft One: https://www.eff.org/deeplinks/2025/07/effs-guide-getting-records-about-axons-ai-generated-police-reports