US Department of Labor AI Guidelines

The US Department of Labor has issued a Field Assistance Bulletin on Artificial Intelligence and Automated Systems in the Workplace under the Fair Labor Standards Act and Other Federal Labor Standards.

The key points in the Bulletin are as follows:

  • The FLSA and other federal labor laws enforced by the Department of Labor’s Wage and Hour Division (WHD) continue to apply regardless of the tools and systems employers use, including AI. Employees are entitled to the protections of these laws even when new technologies are implemented.
  • With regards to hours worked, AI systems used to track productivity metrics or monitor breaks do not substitute for analyzing whether the employee was “suffered or permitted” to work during that time under the FLSA. Automated systems may undercount hours worked, especially during break times, waiting periods or when work is performed at multiple locations.
  • For calculating wages owed, employers using AI to determine pay rates must ensure employees are still paid applicable minimum wage and overtime premium pay based on their regular rate of pay for all hours worked in accordance with the FLSA.
  • AI systems used to process Family and Medical Leave Act leave requests or certifications must be overseen to avoid eligibility and entitlement errors or improper medical inquiries. Such systems also cannot be used to interfere with FMLA rights or retaliate against employees for taking protected leave.
  • Under the PUMP Act amendment to the FLSA, automated scheduling and monitoring systems cannot penalize nursing mothers for taking needed lactation breaks.
  • An employer’s use of AI as a “lie detector test” may be prohibited under the Employee Polygraph Protection Act with limited exceptions.
  • Using AI to surveil employees and take adverse actions for engaging in activities protected by WHD-enforced laws, like filing a complaint, would constitute unlawful retaliation.
  • While AI has the potential to improve compliance when used responsibly, employers are ultimately liable for violations of WHD-enforced laws resulting from the use of these technologies and must ensure responsible use and oversight. The risk of systemic violations across the workforce is heightened with automated systems.

In summary, the Field Assistance Bulletin emphasizes that employers cannot abdicate their responsibilities under federal labor standards simply by adopting AI technologies and must exercise proper human oversight of these systems to maintain compliance.