State Office of the Chief Information Officer. The Consolidated Technology Services Agency, also known as Washington Technology Services (WaTech), supports state agencies as a centralized provider and procurer of information technology (IT) services. Within WaTech, the Office of the Chief Information Officer (OCIO) has primary duties related to IT for state government such as establishing statewide enterprise architecture and standards. 

Office of Equity. The Washington State Office of Equity was established in 2020 within the Office of the Governor to promote access to equitable opportunities and resources that reduce disparities, and improve outcomes statewide across state government.

Washington's Law Against Discrimination. Washington's Law Against Discrimination (WLAD) prohibits discrimination based on race, creed, color, national origin, sex, marital status, and other enumerated factors. Discriminatory practices are prohibited in employment, commerce, credit and insurance transactions, access to public places, and real property transactions

Automated Decision-making Systems Work Group. In the 2021-23 biennial operating budget, the Legislature directed OCIO to convene a work group to develop recommendations for changes in state law and policy regarding the development, procurement, and use of automated decision-making systems by public agencies. A report was submitted on December 1, 2021.

Governor's Directive. In 2022, the Governor directed WaTech to develop guidance for agencies' procurement and use of automated decision systems (ADS) and to conduct a preliminary inventory of these systems currently in use by November 30, 2023.

Office of the Chief Information Officer. The OCIO must:

• by January 1, 2023, in consultation with the Office of Equity, adopt guidance for agencies regarding the development, procurement, and use of an ADS by a public agency and develop guidance for agencies to use when prioritizing analysis of an ADS;
• beginning December 1, 2023, and updated not less that quarterly, publish on its public website an inventory of all reports that have been proposed for or are being used, developed, or procured by public agencies; and
• beginning January 1, 2024, publish on its public website metrics on all approvals, conditional approvals, or denials of agency reports.

The OCIO and agencies, in consultation with the Office of Equity, should adopt interim and long-term prioritization frameworks for allocating resources to address an existing and future ADS and to address any compliance deficiencies.

The OCIO must post each algorithmic accountability report (report) on its public website and invite public comment for a period of no less than 30 days. 

Algorithmic Accountability Review Board. The Algorithmic Accountability Board (board) is created within the OCIO. Membership include legislators, the director of the OCIO, and representatives from state agencies, consumer protection organizations, and civil rights organizations or advocacy organizations representing individuals or protected classes of historically marginalized communities.
 
After receiving public comment on a report, the board must review the report and comments to determine whether the agency's report demonstrates that the ADS meets minimum standards.

The board must:

• annually report on agency compliance with the report requirement and any extensions granted;
• for an ADS implemented prior to January 1, 2026, conduct selective audits of reports and make a finding whether the report demonstrates the ADS meets the minimum standards; and
• beginning January 1, 2026, conduct an annual review of agency audits and compile the information into a report.

Public Agencies. As part of the procurement process, agencies should assess a newly procured ADS. An ADS currently used by the state that produces legal effects on an identifiable natural person should be assessed according to the prioritization framework.

Agencies should provide transparency of use, procurement, and development of an ADS, including ongoing monitoring or auditing, and training to state employees who develop, procure, operate, or use an ADS as to the risk of automation bias.

An agency must:

• give clear notice of use to the people impacted by the ADS;
• ensure the ADS and data used to develop it are made freely available by the vendor before, during, and after deployment for agency or independent testing or auditing;
• ensure any decision made or informed by the ADS is subject to appeal; and
• ensure it can explain the basis for its decision to any impacted person.

Beginning January 1, 2026, agencies must conduct an annual audit on each ADS that has legal effects on people to ensure they do not have differential effects on subpopulations that result over time, and report any findings to the board.

Algorithmic Accountability Report. By January 1, 2026, agencies must complete a report that meets specified requirements for each existing ADS. If an agency does not meet this reporting deadline or receive an extension by the board, the agency must cease use of the unevaluated ADS.

Each report must include specified statements such as a description of the ADS's general capabilities, a description of the purpose and proposed use, the type or types of data input used, and whether there was an examination of potential inaccuracies.

An agency intending to procure or newly develop an ADS for use before January 1, 2026, must file a report with the OCIO at least one month prior to procurement or implementation of an internally developed system. An agency intending to procure or develop an ADS after January 1, 2026, must submit a report and receive a finding of whether the ADS meets minimum standards. For an ADS that involves high risk, the report must also include an evaluation of accuracy and bias by a qualified independent third party.

Enforcement. Any person injured by a material violation of this chapter may institute proceedings against the public agency deploying the ADS in a court of competent jurisdiction for injunctive or declaratory relief, or both, to compel compliance. All relief available in law and equity with respect to the WLAD, and in either event if successful, shall be entitled to recover reasonable attorneys' fees and costs.

Washington's Law Against Discrimination. WLAD is expanded to prohibit discrimination by ADS against any individual on the basis of one or more factors enumerated in the WLAD.

PRO: This is a crucial topic that Washington State is already behind on. ADS are very complex systems that take data from other less complex systems down the chain. Each step of ADS systems should be analyzed. When knowledge surpasses wisdom, we lose control over how that knowledge is being used. This bill aims to invoke wisdom into this process. Government agencies are increasingly using ADS to make critical decisions about people lives'. People need to be able to seek recourse if they are harmed by these decisions. There are systems being deployed without public notice or oversight. Washington has the opportunity to innovate with accountability and transparency. Only transparency can expose harmful uses of algorithms. Agencies could implement an ADS without showing proof that the algorithm works. They also don’t have expert or stakeholder input; there is no accountability. This bill aligns with the work group's recommendations and global trends and is a critical step towards protecting Washington residents.

OTHER: There is a desire to have clarity on what exactly is or is not included in the scope of an ADS. Some systems law enforcement use might be unintentionally captured in the definition.