When police agencies implement new technology, especially that which uses artificial intelligence (AI), it is quite easy to miss the unintended consequences of using such a new, and powerful, addition to policing’s technological “toolbox” for controlling crime and disorder. The evolution of AI is happening so fast, and is so pervasive, that police leaders can easily purchase technology without fully appreciating that vendors are baking some form of AI into their products and services. If the vendor doesn’t disclose this agencies could inadvertently acquire technology with built-in bias. Accordingly, the COP-AI staff have put together a brief AI Implementation Checklist to help police leaders begin the process of vetting technology they’ve acquired, or may acquire, to better understand the future implications of what may initially be perceived as helpful and effective technology.

This checklist is intended to provide some broad guidelines on the issues policing leaders should consider before they implement a AI-centric technology. No matter how large or small the project, police leaders will be well-served by remembering that AI has the potential to advance the police mission or dramatically harm the police-community relationship. This checklist is not intended to provide guidance on every aspect of a technology project involving AI. Rather, it creates a broad set of things to consider prior to implementing any AI-centric technology project. It is intended to help policing organizations enhance community trust and confidence in the police by being transparent, accountable and demonstrating a sincere interest in community feedback and concerns. Click here to access the checklist.