Microsoft has taken a firm stance against the use of generative AI for facial recognition by U.S. police departments. The company’s Azure OpenAI Service, a fully managed, enterprise-focused wrapper around OpenAI technology, will no longer be available for such purposes. Here’s what you need to know:

The Ban

Microsoft’s updated terms of service for Azure OpenAI Service now explicitly prohibit integrations with the service being used “by or for” police departments in the U.S. for facial recognition. This includes any current and potential future image-analyzing models provided by OpenAI. Additionally, the ban extends to “any law enforcement globally,” preventing the use of real-time facial recognition technology on mobile cameras (such as body cameras and dashcams) in uncontrolled environments.

The Context

The recent changes come in the wake of a product announcement by Axon, a company that produces technology and weapons for military and law enforcement. Axon’s new product leverages OpenAI’s GPT-4 generative text model to summarize audio from body cameras. However, critics have raised concerns about potential pitfalls, including hallucinations (even the best generative AI models can invent facts) and racial biases introduced from training data.

Wiggle Room

While the ban applies specifically to U.S. police departments, it does not cover international law enforcement agencies. Additionally, facial recognition performed with stationary cameras in controlled environments (such as back offices) is not explicitly prohibited, although any use of facial recognition by U.S. police is still restricted.

Backdrop of Microsoft’s Approach

Microsoft’s approach to AI-related law enforcement and defense contracts has been nuanced. While they’ve banned certain uses, they’ve also collaborated with OpenAI on projects involving the Pentagon and the Department of Defense. Azure OpenAI Service, which became available in Microsoft’s Azure Government product, includes compliance and management features tailored for government agencies, including law enforcement.

In summary, Microsoft’s ban on using Azure OpenAI Service for facial recognition by U.S. police departments reflects a growing awareness of the ethical and practical implications of AI technology. As businesses and governments grapple with these issues, transparency, accountability, and responsible use remain critical considerations.