Home / Technology / AI Hiring Lawsuit: Who's Liable for Bias?
AI Hiring Lawsuit: Who's Liable for Bias?
21 Sep
Summary
- A lawsuit questions responsibility for AI hiring tool discrimination.
- Plaintiffs allege Workday's AI perpetuates past discrimination patterns.
- The case could set a precedent for AI software companies and clients.
A significant legal battle is underway as a lawsuit challenges who is responsible when AI hiring tools allegedly discriminate. A group of job seekers has asked a federal judge in San Francisco to certify their case against human resources software provider Workday as a class action. The plaintiffs contend that Workday's AI hiring tools, while operating as designed, perpetuate patterns of past discrimination, leading to the rejection of Black and other applicants. This case is particularly noteworthy as it involves older AI technology, predating the current generative AI surge.
Workday denies these allegations, asserting its AI tools focus solely on job qualifications and not protected traits. The company argues the plaintiffs cannot meet the burden for a class action. The outcome of this litigation could have far-reaching implications, given that Workday's software is used by two-thirds of the Fortune 500, and over 80% of US employers utilize AI in their hiring processes. Employment litigators view this case as critical, potentially creating a roadmap for future claims against both AI software creators and their users.
The lawsuit began in 2023 after co-lead plaintiff Derek Mobley experienced repeated rejections for various jobs. His lawyers are targeting Workday directly, rather than the companies using the software. While Workday states its recruiting software is customizable by clients, the plaintiffs claim the software itself contributes to discriminatory outcomes by using existing workforce data for training, thereby embedding existing biases. In 2024, a judge allowed the case to proceed, finding that the plaintiffs plausibly alleged a disparate impact, a legal theory where neutral practices disproportionately affect a protected group without business necessity.
The next major hurdle is obtaining class certification. The plaintiffs aim to represent subclasses for African Americans, women, individuals over 40, and those with disabilities who applied for jobs and faced the alleged discriminatory screening. In 2024 alone, Workday Recruiting processed over 356 million applications. The core challenge for class certification lies in demonstrating that the alleged discrimination stems from common features of Workday's technology rather than individualized hiring decisions made by its numerous clients. A hearing on class certification is scheduled for March 9, 2027.
