San Jose trained a seventh of its workforce in AI by building the course itself
The reporting says more than 1,000 city employees have finished an in-house program run with San José State University since 2024, though what is measured so far is enrollment rather than results.
This is not our reporting
This story was reported and published by Route Fifty. What follows is our summary of it, and why it matters for people writing policy.
Read the original at Route Fifty ↗Our summary
Rather than buy a vendor curriculum, San Jose built its AI training in house: self-paced online modules plus a 10-week course developed with San José State University, covering responsible use, prompt engineering and building custom tools for staff's own jobs. According to the piece, more than 1,000 employees — about 15 percent of a roughly 7,000-person workforce — have completed it since the 2024 launch, sessions are typically oversubscribed, and the city is aiming for 30 percent participation by June 2027. Be clear about what that measures: uptake, not outcomes. The reporting cites no evaluation of time saved, service quality or cost avoided.
Read the full story at Route Fifty ↗
More from The Effective
California built a public tracker for AI job loss and so far it finds almost none
The state's unemployment agency and a university lab match jobless claims against AI task-exposure scores, and the reporting says the only moving line is claims from college-educated workers in high-exposure jobs.
AI screening of permit applications cut Honolulu's median housing permit wait by about 40 percent
Cities are putting automated completeness checks in front of human plan reviewers, and the reporting finds real time savings in Honolulu alongside a pilot Seattle judged not ready to adopt.
Early state pilots suggest agentic AI can speed up permits and prescriptions
A new playbook catalogs early agentic-AI experiments in Virginia, Utah, and Boston, including one pilot state officials say found $1.4 billion in annual savings and cut permit times by nearly 80%.