San Jose Mayor Faces Backlash Over AI Spending
San Jose Mayor Matt Mahan has found himself at the center of controversy after spending over $35,000 on AI tools for city workers. The funds were used to purchase 89 ChatGPT licenses, costing $400 per account, for employees to use in their daily tasks. This move is part of a broader initiative aimed at integrating artificial intelligence into city operations.
By next year, the city plans to train approximately 1,000 workers—about 15% of its total workforce—in the use of AI tools. These tools are expected to assist with various tasks such as responding to pothole complaints, managing bus routes, and utilizing vehicle-tracking surveillance cameras to solve crimes. Even some of Mahan’s staff have used AI to draft talking points before events, while the mayor himself relied on it to help write a $5.6 billion budget for the upcoming fiscal year.
Mahan is pushing for a growing number of the nearly 7,000 government workers in Silicon Valley’s largest city to embrace AI technology. He emphasizes the importance of experimentation and transparency, stating that the goal is to identify problems, flag them, and collaborate across agencies to find solutions. “It’s always bumpy with new technologies,” he said in an interview. According to Mahan, adopting AI tools will reduce mundane tasks and improve the city’s ability to serve its one million residents.
However, not everyone is pleased with the mayor’s approach. Some critics argue that the investment in AI is a distraction from more pressing issues like public safety, housing, and community pride. One commenter on social media suggested, “Here’s a real idea for AI that works: Replace Matt Mahan with AI.” Others have questioned the value of the technology, noting that the results of AI usage in the city are not visible to taxpayers.
Despite the criticism, some city employees have found success with AI. Andrea Arjona Amador, who leads electric mobility programs at the city’s transportation department, used ChatGPT to secure a $12 million grant for electric vehicle chargers. She also created a customized AI agent to manage correspondence related to grant proposals and helped draft a 20-page document. Arjona Amador noted that the previous process involved long hours and weekends to complete grants, but AI has significantly reduced the workload.
The mayor’s push for AI adoption aligns with similar efforts in other Bay Area cities. San Francisco Mayor Daniel Lurie recently announced a plan to provide nearly 30,000 city workers, including nurses and social workers, with access to Microsoft’s Copilot chatbot. The plan includes robust privacy and bias safeguards, ensuring that technology enhances—rather than replaces—human judgment.
While San Jose has implemented similar guidelines and hasn’t reported major issues with its pilot projects, concerns about AI’s reliability persist. For instance, ChatGPT’s digital fingerprints were found on an error-filled document published by the U.S. Health Secretary Robert Kennedy Jr.’s commission. In Fresno, California, a school official was forced to resign after trusting an AI chatbot that fabricated information in a document.
OpenAI introduced a new product called Operator, which allows AI agents to perform tasks beyond text generation, such as accessing computer systems and scheduling calendars. However, developing and selling these agents remains a key focus for the tech industry. Despite the potential, market research group Gartner predicts that over 40% of agentic AI projects will be canceled by the end of 2027 due to costs, unclear business value, or inadequate risk controls.
Mahan remains optimistic about the potential of AI tools to streamline bureaucratic processes. He believes that roles in finance, accounting, HR, and grant writing can see productivity gains of 20 to 50%. “There’s just an amazing amount of bureaucracy that large organizations have to have,” he said. “These are the kinds of roles where we think our employees can be more productive quickly.”