DC becomes first major US city to require AI training

19 February 2026

by Jonathan Andrews

Washington DC has become the first major US city to introduce mandatory responsible AI training for all government employees and contractors, embedding AI literacy across its public workforce.

Announced by the mayor, Muriel Bowser, the requirement builds on Mayor’s Order 2024-028, signed in February 2024, which established the city’s governance framework for artificial intelligence. The Order requires agencies to demonstrate that any proposed AI tools align with six values: clear benefit to residents, safety and equity, accountability, transparency, sustainability, and privacy and cybersecurity.

The new training, delivered in partnership with InnovateUS, consists of two self-paced online courses that take less than two hours in total. All employees and contractors must complete them within 90 days.

In an interview with Cities Today, Stephen Miller (pictured), DC Chief Technology Officer and Director of the Office of the Chief Technology Officer, said the mandate reinforces an existing governance structure rather than operating as a standalone initiative.

“Making this mandatory for everyone sends a clear message: responsible AI use is everyone’s responsibility,” he said. “Just as we expect all employees to understand cybersecurity basics, data privacy requirements, and ethical conduct standards, we now expect everyone to understand responsible AI use.”

Embedding oversight into practice

To determine whether the training is influencing behaviour, DC will rely on multiple oversight channels already embedded in its AI framework.

The AI Taskforce, led by OCTO, tracks agency requests for AI tools and oversees their implementation. As staff complete the training, the Taskforce will assess whether proposals demonstrate clearer alignment with the six AI Values and whether agencies are applying “humans in the loop” principles in practice.

Community feedback is another pillar of evaluation. Through the AI Values Alignment Advisory Group, which brings together residents and external stakeholders, the District will gather insight into whether AI-enabled services are improving delivery and maintaining public trust.

Monitoring will also extend to how employees engage with approved enterprise tools inside the District’s secure Microsoft 365 environment. Rather than focusing solely on completion rates, officials will examine usage patterns to understand how responsibly tools such as Microsoft Copilot Chat are being used in day-to-day operations.

“Ultimately, behaviour change in responsible AI use is an ongoing process, not a one-time event,” Miller said. “This training–which takes less than two hours total to complete–establishes a shared baseline, and we’ll continue to refine our approach based on real-world outcomes and feedback from both employees and the communities they serve.”

Linking training to procurement

Completion will be tracked through a learning management system integrated with DC Government’s Single Sign-On credentials. Access to AI tools will be restricted for employees or contractors who do not complete the training within the 90-day window.

The requirement is closely aligned with procurement and tool approval processes. Agencies seeking to deploy new AI systems for service delivery or internal operations must undergo formal vetting by the AI Taskforce to demonstrate alignment with the six AI Values.

“The training and our tool approval processes work hand-in-hand,” Miller said. “When an agency requests a new AI tool for service delivery or operations, it must go through formal vetting by our AI Taskforce to demonstrate alignment with our six AI Values–that’s required under Mayor’s Order 2024-028.”

For Miller, the mandate is less about course completion and more about embedding standards of conduct into everyday use of AI across government.

“The real accountability isn’t just about checking a box–it’s about ensuring that when DC Government employees use AI tools, they’re doing so in ways that protect resident privacy, maintain transparency, and deliver genuine benefits to the communities we serve.”

Image: DC Office of the Chief Technology Officer

https://cities-today.com/wp-content/uploads/2025/01/dreamstime_m_158418624-image.jpg

How collaboration can improve micromobility within cities