US workers are fighting to protect their jobs from AI – and winning

As governments adopt artificial intelligence (AI), public-sector workers are gaining ground in how it's used and securing safeguards against its potential harms. In Oakland, California, a city union overwhelmingly voted to authorize a strike on 22 October to ensure workers won’t be laid off or disciplined by AI. Union representative Julian Ware said, "We just want the city administration to really be intentional about how it’s engaging with AI products."

The city’s tentative, three-year agreement includes protections such as retraining or reassigning workers displaced by AI rather than firing them. The strike has been put on hold due to this agreement. Nationwide, public-sector labor groups are seeking explicit protections against AI and automation, fearing worsened working conditions and job threats.

Since ChatGPT's debut in 2022, AI and automation technologies have infiltrated multiple sectors, including public-sector workplaces. Nearly 70% of state and local government agencies have adopted or are implementing generative AI, according to a survey by Ernst and Young.

Workers are organizing and winning protections through contract negotiations, as policymakers have moved too slowly to address AI's potential effects on workers. Liz Shuler, AFL-CIO president, argues that unions must institute guardrails to prevent job losses due to AI. Some unions have drawn lessons from private-sector battles, such as the Communications Workers of America's recommendation for AI protections, which led to some safeguards for California workers in October 2026.

Public-sector employers are increasingly deploying AI tools to monitor and supervise employees, known as "bossware." These tools can rank and evaluate workers, track location and activity, monitor productivity, scan biometrics, use cameras and wearable devices, and predict emotions. However, humans should always have control over AI systems, providing oversight, argues Lorena Gonzalez, California Federation of Labor Unions president.

Automation technologies could aid public-sector workers, but the ultimate goal is to ensure that humans are in charge, providing judgment and skill to override the system when necessary. Employers often deploy automation technologies during budget constraints, but unions have been pushing for AI protections since the late 2010s, with battles intensifying as automation advances and generative AI becomes more common.