AI Internal Revolution in San Jose: Why City Governments Need to Cultivate Their Own AI Capabilities

In the San Jose, California Department of Transportation, a landscape construction plan submitted by a contractor once required more than five hours of manual review, back-and-forth emails, and revisions. Today, a construction inspector named Amanda Nichols, with the help of an AI assistant she built herself, has compressed the entire process to within a single day—usually requiring only one or two minor adjustments for approval. This tool, which she calls the "Landscape Inspection Assistant," is not a product from a tech company but was developed by her after participating in the city government's internal AI skills enhancement program.

Since its launch in 2025, San Jose’s program has trained over 1,000 city employees. Unlike common AI training on the market, this program, jointly designed by the city government and San Jose State University, lasts 10 weeks. The focus is not on teaching AI principles but on having employees directly build AI tools that address their specific daily work pain points. Participants come from different departments such as IT, transportation, and fire services, and many had no prior experience with AI before joining.

This "internal building" model is redefining the path of government digitalization. Traditional smart city projects often rely on procuring large-scale AI systems from external vendors, but San Jose has chosen a decentralized path: distributing AI capabilities to every employee, allowing them to become the automation designers of their own workflows.

The intersection of urban planning and AI is often understood as "autonomous traffic signals" or "crime prediction algorithms," but San Jose's practice reveals a more grassroots transformation—AI is entering the daily paperwork, approval processes, and project management of city hall. Project administrator Paulina Hen used an AI assistant to reduce the time for drafting project charters from months to a single day; fire department employees developed tools to verify that response vehicles are fully equipped; the IT department built AI assistants to support Power BI dashboards. These tools do not seek to be flashy; they only solve specific efficiency issues.

From the perspective of urban governance, this change holds systemic significance. First, it lowers the barrier to AI application. When employees can create AI tools without relying on professional IT teams or external vendors, the agility of city services significantly improves. Second, it changes the relationship between government and technology—cities are no longer passive consumers of technology but active producers. San Jose has shared the course blueprint with the GovAI Alliance for other cities to reference, effectively building an intergovernmental AI collaboration network.

But this path also faces challenges. How can employee-developed AI tools ensure security, compliance, and fairness? Paulina Hen emphasizes that human oversight and privacy awareness are the bottom line. Amanda Nichols points out that the best AI solutions are often not the most complex but those that precisely solve everyday problems. This suggests that cities need to establish an internal AI governance framework that empowers employee creativity while managing risks.San Jose's approach is not an isolated case. Globally, city governments are shifting from "procuring AI" to "cultivating AI literacy." Singapore has launched its "Smart Nation" AI training program, and Barcelona has established a citizen AI lab. But what sets San Jose apart is the scale of its "internal build"—involving over 1,000 employees, spanning multiple departments, with tools directly embedded into workflows.

In the long run, this internal revolution may reshape the operational logic of urban digital infrastructure. When every department has its own AI assistant, the city's operating system is no longer a single central brain but a distributed network of countless micro-intelligent nodes. Beyond efficiency gains, this model also brings benefits in talent retention and public trust—citizens see their tax dollars used to tangibly improve service speed, rather than being spent on flashy but impractical systems.

San Jose's story tells us that the competition among future cities will depend not only on computing power or data volume, but also on whether governments have the ability to internalize AI as an organizational instinct. When cities learn to "build their own wheels" instead of always buying off-the-shelf solutions, digital governance truly begins to permeate every detail of public service.