San Francisco rolls out Microsoft's AI tools to 30,000 govt workers

With Microsoft 365 Copilot, the US city aims to cut down on administrative burden, improve public services, and become a benchmark for responsible AI use in government

The city of San Francisco, on the west coast of the United States, has announced the rollout of AI-powered tools to nearly 30,000 public sector employees. This marks one of the largest and most ambitious government-level adoptions of artificial intelligence (AI) in the world.

 

San Francisco’s Mayor, Daniel Lurie, confirmed on Monday (local time) that employees across all city departments, including healthcare workers, social service staff, and administrative personnel, will now have access to Microsoft 365 Copilot Chat, an AI assistant powered by OpenAI’s GPT-4o model.

 

This tool will help reduce time spent on routine administrative tasks like data analysis, drafting reports, and translating content into over 40 languages. With this move, San Francisco aims to boost efficiency and allow workers to spend more time directly serving residents.

 

“San Francisco is the global home of AI, and now we’re putting that innovation to work,” said Mayor Lurie. “We’re showing the world how governments can use technology to deliver better results.”

 

What is Microsoft Copilot Chat?

Microsoft 365 Copilot Chat is a smart assistant built into popular workplace tools such as Word, Excel, and Outlook. With the help of AI, it can draft emails, summarise long documents, and even help translate languages. These are all repetitive tasks that often consume hours of government workers’ time. 

 

AI tools helped workers save up to five hours a week, the city found during its six-month pilot programme with 2,000 employees. It was especially effective in 311 public service lines (similar to helplines in India), improving response times for complaints related to waste management, homelessness, and multilingual service needs.

 

The city is now launching a five-week training programme for employees. This includes live workshops, public sector-specific courses, and support from InnovateUS, a nonprofit that focuses on preparing government staff for digital transformation.

 

Security and ethics top priorities in AI adoption

The platform will run on Microsoft’s secure government cloud, with strict data protection measures. Updated AI guidelines, released on July 8, ensure that all AI use remains transparent and responsible. Every department must disclose how it is using AI, and staff remain accountable for any AI-generated content they share.

 

“AI is here, and San Francisco is meeting the moment,” said Michael Makstman, the city’s Chief Information Officer. “We’re not just talking about innovation—we’re doing it responsibly and at scale.”

 

How is the govt adapting artificial intelligence?

San Francisco is also applying AI in other areas like:

- Improving traffic safety through predictive analytics

- Connecting homeless individuals with services faster

- Simplifying permit approvals, similar to India’s building and trade license processes

 

The city plans to build on feedback from employees to improve AI tools over time. It hopes to set an example for governments leveraging AI. “As our city and the world embrace AI technology, San Francisco is setting the standard for how local government can responsibly do the same,” Mayor Lurie said.

 

OpenAI offer custom AI

This development also comes at a time when OpenAI, the creator of GPT-4o, has begun offering highly customised AI services starting at $10 million, with early clients including the US Department of Defence and Southeast Asia's super app Grab. As part of this strategy, OpenAI is embedding forward-deployed engineers (within organisations to tailor GPT-4o to specific needs, ranging from workflow automation to advanced chatbots. This shift places OpenAI in direct competition with established consulting giants like Palantir and Accenture.