VeriPol applied natural language analysis to the written text of a robbery report -- what a victim or witness told police -- to flag characteristics statistically associated with false reports. This makes it a materially different kind of police AI from the video and biometric systems that make up most of this catalogue: it analysed language in a written account rather than an image, video, or biometric sample.

Launched nationally in October 2018 following a 2017 pilot in Malaga and Murcia, and promoted at the time as over 90% accurate. The evidence base was thin from the outset: the model was trained on 1,122 robbery reports from 2015, built on a starting assumption that 57% of violent robbery reports are false, and a University of Valencia analysis identified serious design deficiencies. Usage fell sharply over time, from roughly 84,000 reports analysed by October 2020 to 3,762 in 2022.

Spain's Dirección General de la Policía confirmed to the transparency organisation Civio that the tool stopped being used on 21 October 2024 -- three months after the EU AI Act came into force, which is widely reported as a factor in the decision, since the Act classifies tools of this kind as high-risk. The Interior Ministry has declined to release which police stations deployed it or how often, and a transparency complaint on that point remains open.