Artificial Intelligence Agreement Between Kaiser Permanente and Healthcare Unions
An MIT Sloan report examines the lessons offered to employers by the artificial intelligence labor agreement made between Kaiser Permanente and unions in 2025.
The artificial intelligence agreement signed in 2025 between Kaiser Permanente and the Coalition of Kaiser Permanente Unions serves as an important guide for employers. A report prepared by the MIT Sloan School of Management highlights the details of this process and the importance of worker participation.
Artificial Intelligence and Its Effects on the Business World
Artificial intelligence helps streamline daily operations for healthcare companies by automating paperwork and improving diagnoses. These technological changes inevitably lead to certain disruptions and concerns for human workers.
Details of the Historic 2025 Agreement
Collective bargaining negotiations took place in 2025 between healthcare giant Kaiser Permanente and unions representing 62,000 nurses, technicians, and healthcare workers. As a result of this process, a comprehensive agreement was reached that integrates worker input into every stage of the development and use of artificial intelligence.
Findings of the MIT Sloan Research Report
MIT Sloan School of Management professors Thomas Kochan, Erin Kelly, and Dr. Arrow Minster from San Francisco State University published a new report. This report documents the ways the two parties reached an agreement while offering important takeaways for employers in other industries.
Negotiation Process and Collaborative Efforts
Preparing for the contract renewal process in 2024, the parties identified artificial intelligence as a central focus. Prior to the formal negotiations in May 2025, a joint conference attended by approximately 380 leaders was held in February.
Governance Strategies and Task Force
Combining high-level oversight with grassroots strategies, a joint Artificial Intelligence Task Force consisting of five worker and five management representatives was established. Additionally, more than 3,500 unit-based teams were tasked with identifying areas to improve workflow.