Palm Desert weighs risk-based approval rules for AI in city work

Palm Desert’s council is set to consider a citywide AI policy that distinguishes routine assistance from uses requiring senior approval. The proposal would give the city’s previously established artificial intelligence officer responsibility for administering governance and risk classification.

For city managers and department leaders, the proposal defines who must approve AI uses that expose sensitive information or create legal or financial commitments. It offers a concrete governance approach for permitting routine staff assistance without treating autonomous actions as equivalent to everyday tool use.

Palm Desert is considering how to let staff use AI for routine work while controlling access to sensitive information and actions taken on the city’s behalf. Its proposed citywide policy would classify uses as low, moderate or high risk based on capabilities, access and potential impact. Moderate- and high-risk uses would require approval from a Department Director or City Manager, including uses involving restricted information or independent communications and actions without prior human approval. The staffing foundation is an artificial intelligence officer position established in March 2026, responsible for administering the policy and risk-classification program; the report provides no budget figures. For other local governments, the practical starting point illustrated here is assigning governance responsibility and defining which uses need escalation. The proposal also prohibits fabricated records, false attribution and bypassing security controls or required approvals. Adoption remains pending; approval would make the policy effective immediately.

Evidence supplied by NBC Palm Springs.