White House Calls for Scrutinizing AI Discrimination in Hiring

Artificial intelligence (AI) is rapidly transforming modern businesses, notably in how they recruit and hire employees. AI tools offer efficiency and speed in sifting through resumes and applications and in curating talent. However, the increased use of these tools has raised concerns about their potential to perpetuate or even intensify discriminatory practices. Addressing this fear, the Biden Administration recently issued an executive order that directs the U.S. Department of Labor to research the possible discriminatory effects of using AI in recruitment and hiring.

The executive order (EO) points to data showing that AI “virtual decision-makers” — which rely on computer algorithms to analyze candidates’ credentials — can reproduce and intensify existing inequities in hiring and recruitment. This is because they may be trained in historical biases within an organization, so that they continue patterns of discrimination and perhaps create new ones. 

Among other measures, the EO instructs the DOL to undertake the following:

1. Disparate impact analysis and worker displacement assessment — The DOL is to examine the potential displacement of workers as a result of AI automation in the workplace. Additionally, it must evaluate the infrastructure and resources that may be available to support individuals who lose their jobs due to AI advancements. 
2. Development of principles and best practices for employers — By April 2024, the DOL is expected to publish a set of guidelines aimed at mitigating potential harm against employees, with an emphasis on preventing discriminatory outcomes. Developing these principles and best practices will involve soliciting input from other agencies, labor unions and other stakeholders.

The Biden Administration’s stated goal is to instill responsible development and use of AI while at the same time preventing harm to workers. While existing legal protections against discrimination remain in place, the EO calls for more specific strictures tailored to the complexities of AI-driven hiring.

A chief problem in policing AI-orchestrated discrimination is that it is not always apparent to applicants whether AI tools are being used. State laws have been proposed in New Jersey and New York that would require employers to notify candidates that AI is being used to screen applications. They would require companies to conduct periodic “bias audits” to assess its predicted compliance with anti-discrimination laws like federal Title VII and the New Jersey Law Against Discrimination. However, these bills are still in committees and have not seen forward movement.

In the meantime, people seeking jobs or promotions should be aware that AI may be used in the evaluation process and that certain criteria specified in an AI formula might inadvertently discriminate. For example, requiring a degree in a recently minted major or vocational concentration may work against older employees who graduated many years earlier, before such degrees were available.

If you believe you are the victim of illegal job discrimination, the attorneys at Deutsch Atkins & Kleinfeldt, P.C. in Hackensack, New Jersey have the experience and resources to fight for your rights. Call 551-245-8894 or contact us online to schedule a confidential consultation.