The Personnel System Reform Act (PSRA) provides for collective bargaining for classified employees of state agencies and institutions of higher education. Matters subject to bargaining under the PSRA include wages, hours, other terms of conditions of employment, and the negotiation of any question arising under a collective bargaining agreement. The PSRA prohibits bargaining over management rights, which, among other things, includes the functions and programs of the employer, use of technology, and the structure of the organization.
Employees of cities, counties, and other political subdivisions of the state, bargain their wages and working conditions under the Public Employee's Collective Bargaining Act ( PECBA). Certain higher education and state employees also bargain under PECBA. For certain higher education employees under PECBA, the use of technology is considered a management right and is a prohibited subject of bargaining. For the rest of the employees and employers covered under PECBA, the use of technology is a permissive subject of bargaining, where the parties are neither required nor prohibited from bargaining the subject matter. Bargaining is required if the use of technology affects a mandatory subject of bargaining, such as wages, hours, or working conditions.
Public employers covered by PECBA and the PSRA are required to bargain over the decision to adopt, or modify current uses of, artificial intelligence (AI) technology if the adoption or modification affects employees' wages or performance evaluations.
The requirement to bargain does not apply to the implementation or modification of AI technology if it is part of an update made by a third party to technology already in use and does not meaningfully impact employee's wages or performance evaluations. A third party is an individual or entity that provides services such as technology services, digital services, equipment, or software, but does not have an employment or a co-employment relationship with the employer.
AI means the use of machine learning and related technologies that use data to train statistical models for the purpose of enabling computer systems to perform tasks normally associated with human intelligence or perception, such as computer vision, speech or natural language processing, translation, decision making, and content generation.
Machine learning is the process by which AI is developed using data and algorithms to draw inferences to automatically adapt or improve its accuracy without explicit programming.
The committee recommended a different version of the bill than what was heard. PRO: The bill allows for bargaining if AI impacts wages or performance evaluations and there is no impact if there is an update to current technology. Technology has changed a lot since the prohibition on bargaining over it was adopted. There are no guidelines for implementation of AI. Workers should have the right to bargain over AI if it replaces their job. The private sector is already using AI for performance evaluations.
PRO: Representative Lisa Parshley, Prime Sponsor; Cherika Carter, WA State Labor Council AFLCIO; Kati Durkin, Washington Federation of State Employees; Christine Reid, IBEW 77.
N/A
The committee recommended a different version of the bill than what was heard. PRO: The bill is good for both business and consumers. Even though it is a management right, slowing down the implementation of AI is the right thing to do. It will improve morale and lead to more employee buy-in. The bill will lead to better informed decision making. Employees should have the ability to bargain over technology that will replace their job. We need to ensure it is a tool to improve, not replace, jobs. AI may produce unreliable results if workers are not consulted.
CON: The bill goes too far and impedes management rights. For interest arbitration employees, it will put the decision to adopt AI into the hands of an arbitrator. Employers must already bargain the impact of these decisions. The bill does more than just change the timing and it is broader than just bargaining the impacts. The definition of AI is overly broad. This will delay and inhibit cost saving decisions. AI is a tool that can improve workplaces and help with efficiencies. The bill upsets the balance of labor and management rights.
OTHER: This is a fundamental shift in management rights. As technology advances, management should have the ability to decide how and when to use certain tools. The parameters of the bill are not clear and it would inhibit the procurement of large IT projects. There is already an AI accountability framework in state government. It is unclear whether the bill applies to AI performing evaluations, AI collected data being used in evaluations, or both. Employers would have to delay program updates to ensure compliance with the bill.
PRO: This bill is essential to provide public sector workers with collective bargaining rights when AI impacts their wages or performance evaluations. This will protect workers from potential job losses and ensure that AI is implemented in a manner that is beneficial to both employers and employees. This does not change existing bargaining practices but instead shifts the timing of negotiations to prevent harm before it occurs. The implementation of AI requires the input of frontline workers to ensure systems are reliable, and the bill ensures that these workers' insights are considered. This encourages collaboration between employers and workers, fostering smoother and more effective integration of new technologies in the workplace.
CON: This is unnecessary, as state and local governments are already required to bargain over wages, hours, and working conditions, and AI should not be treated as a special subject for bargaining. This creates confusion with vague definitions of AI and unnecessary new bargaining requirements that could impede state and local governments' ability to adopt new technologies effectively. The fiscal impact is a concern, as the bill could add significant costs, especially with multiple bargaining units and AI implementations that could involve multiple sessions. Mandating bargaining over AI decisions infringes on management rights and could delay the implementation of cost-saving and innovative technologies, particularly in areas like wildfire prevention and other critical services.
OTHER: This bill raises concerns for the executive branch of state government regarding the bill's potential impact on cybersecurity controls, including network monitoring data that could be used for both cyber threat prevention and performance evaluations. The bill may hinder the proactive detection and prevention of cyber threats by delaying the adoption of necessary tools. Employee bargaining units should be involved in decisions about AI implementation, a more balanced approach, such as providing 180 days' notice before any decisions are made, to allow for adequate input while mitigating risks to security are preferred. The use of technology should remain a management right.