Microsoft Bans U.S. Police Use of AI for Facial Recognition

Microsoft has reinforced its prohibition on the use of generative AI for facial recognition by U.S. police departments through Azure OpenAI Service.

The updated language, introduced on Wednesday, explicitly bars integrations with Azure OpenAI Service from being utilized by or for police departments in the U.S. for facial recognition purposes.

This restriction encompasses both current and potential future image-analyzing models developed by OpenAI.

Moreover, a new provision extends the ban to encompass "any law enforcement globally," explicitly forbidding the utilization of real-time facial recognition technology on mobile cameras in uncontrolled environments such as body cameras and dashcams.

These policy adjustments follow closely on the heels of Axon's unveiling of a new product utilizing OpenAI's GPT-4 generative text model to summarize audio from body cameras.

Concerns raised by critics regarding potential pitfalls, including hallucinations and racial biases, have underscored the need for such regulatory measures.

Although it remains uncertain whether Axon utilized GPT-4 via Azure OpenAI Service, speculation persists regarding the influence of Axon's product launch on Microsoft's policy update.

OpenAI had previously imposed restrictions on the use of its models for facial recognition through its APIs.

Notably, the revised terms allow for some flexibility. While a complete ban on Azure OpenAI Service usage applies solely to U.S. police departments, it does not extend to international law enforcement.

Additionally, the ban does not encompass facial recognition conducted with stationary cameras in controlled environments.

This aligns with Microsoft's and its close partner OpenAI's evolving approach to AI-related engagements with law enforcement and defence sectors.

Recent collaborations with entities such as the Pentagon underscore a shift from previous limitations on providing AI technology to military applications.

Since its integration into Microsoft's Azure Government product in February, Azure OpenAI Service has introduced enhanced compliance and management features tailored to government agencies, including law enforcement.

Candice Ling, SVP of Microsoft's government-focused division Microsoft Federal, has affirmed the commitment to seeking additional authorization from the Department of Defense (DoD) for Azure OpenAI Service workloads supporting DoD missions.

In an update following the initial publication, Microsoft clarified that the ban on facial recognition applies solely to the U.S. and is not a blanket prohibition on police departments utilizing the service.