The past few years have seen a flurry of government and court activity related to the use of artificial intelligence in the workplace, and we’ve brought it all together in one place. This insight reviews all laws, regulations, guidance documents, and litigation that impact employers and the use of AI.
State and local workplace laws
We are beginning to see a patchwork of different state and local laws regulating the use of AI in the workplace.
Colorado is the first state to enact a law that prohibits employers from using AI to discriminate against employees and requires companies to take extensive steps to avoid such algorithmic discrimination. It became a state. This landmark AI law also imposes extensive rules on developers of risky AI systems and the companies that use them. (May 2024) Illinois introduces AI workplace bill that would require employers to notify applicants and workers if they use AI for hiring, discipline, termination, or other workplace-related purposes. It became the second state to pass the . The law would also prohibit employers from using AI in ways that result in workplace discrimination. (September 2024) New York City’s Local Law 144 is the first law in the nation to impose obligations on employers when AI is used for employment purposes (including mandatory bias audits), but it also It is only triggered when it plays a major role in the decision. However, this law is considered fairly ineffective, as employers can argue that it does not apply if they ensure that human managers play a major role in the decision-making process. (July 2023) What’s Next? More than 30 states have established AI commissions or task forces and have begun issuing reports and recommendations, many of which will become legislation. California and New York are expected to become frontline states by 2025. California lawmakers reneged on a chance to pass a landmark AI discrimination bill in 2024 that would require employers to notify and potentially provide accommodations to workers. If AI is used in a particular important way during recruitment or employment, but legislators are unable to agree on a final version by important deadlines. California also came tantalizingly close to passing legislation that would require developers of high-risk AI models to conduct safety testing and implement shutdown mechanisms to prevent serious harm, but Governor Newsom announced in September vetoed this. Meanwhile, the New York State Legislature unsuccessfully attempted to pass a statewide law in 2024 that would address perceived flaws in New York City law.
federal law
What have D.C. lawmakers been doing regarding the use of AI in the workplace?
No federal AI workplace law – There are no federal laws yet that specifically regulate the use of AI in the workplace. However, the Equal Employment Opportunity Commission (EEOC) will warn employers as early as 2022 that using AI to evaluate job applicants and employees may violate the Americans with Disabilities Act (ADA). uttered. Followed by the release of a technical assistance document in 2023 that applies long-standing legal principles to uncover potential Title VII violations when employers use AI to support employment and employment-related activities. Then he said. Failed Privacy Law – An early version of the proposed U.S. Privacy Rights Act would require employers to notify applicants and workers when AI is used in workplace decision-making, and workers would be required to notify them of the consequences. employees had the right to opt out of the use of AI in employment decisions. However, the proposal was amended to remove these provisions before it was ultimately rejected by Congress. Given the change in leadership in Congress, we do not expect similar proposals to gain traction in the near future. Other federal proposals that failed – Last year also saw several other proposals in Congress related to regulating AI in the workplace, from the No Robot Bosses Act to the Algorithmic Liability Act. None have made much progress, and similar proposals are not expected to come close to passing in 2025. What’s next? As mentioned above, we do not expect a Republican-controlled Congress to enact workplace-related AI legislation in 2025 or 2026. Instead, the issue will continue to be active at the state and local level.
Congressional Recommendation
Although Congress did not pass any AI-related workplace legislation, leadership issued a non-binding report with recommendations for future actions.
U.S. House Bipartisan AI Task Force Report – This landmark report outlines the key challenges and opportunities for AI adoption across sectors, and provides recommendations to help employers stay ahead of expected guardrails. We provide. (December 2024) Senate AI Roadmap – A bipartisan group of Senate leaders releases potential legislation that would govern the use of AI across the country and ensure the United States remains at the forefront of AI innovation. We announced an AI roadmap that shows our expectations for the future. (May 2024) What’s next? Task forces, committees, and caucuses will continue to present recommendations to the full Congress, but private technical leaders will have more of a role in shaping proposals. expected to play a major role.
execution action
The White House has been active in AI in recent years.
Biden Executive Order on AI – President Biden’s 2023 Executive Order is the federal government’s most ambitious effort to date to corral AI. Although no new laws affecting the private sector were enacted, it did prompt federal agencies to shape their regulatory approaches to AI. (October 2023) White House Blueprint for AI Bill of Rights – But President Biden’s first action on AI and the workplace wasn’t an executive order. The previous year, the White House issued a nonbinding white paper aimed at supporting the development of policies and practices that protect civil rights in the construction, deployment, and governance of automated systems. (December 2022) White House AI Fact Sheet for Employers – The Biden administration also provided employers with a set of best practices when using AI for workplace purposes. Although it does not carry the weight of law, it may be relied upon by courts and other bodies in the early stages of AI risk management and workplace litigation. (May 2024) What’s next? Donald Trump appointed David Sachs as America’s first AI czar, demonstrating a focus on innovation and deferential regulation, and Biden This is a sure sign that the AI presidential order will be repealed. We may also see a new DOGE initiative that deploys AI to streamline government spending.
Federal agency publications
Various federal agencies have issued guidance and technical assistance documents to clarify how regulators should treat AI.
Court actions and discrimination claims
Workers and labor advocacy groups are beginning to take action against employers for AI-related misconduct.
Lawsuits Against AI Workplace Screening Tools – A California federal court has dismissed an employment discrimination lawsuit against an AI-based vendor by more than 100 employers using the vendor’s screening tools, allowing disgruntled job applicants to allowed employment discrimination lawsuits to be brought against base vendors. The judge’s decision means that Workday is acting as a “representative” for all the employers that rejected him, and that it has banned customers based on the theory that its algorithmic screening tools are biased against his race. The class action lawsuit against Workday was allowed to continue based on employment decisions made by Workday. such as age and presence of disabilities. (July 2024) FTC investigates AI hiring tools – ACLU asks Federal Trade Commission to investigate AON’s personality assessment tests, video interview tools, and cognitive ability assessment screening devices, all powered by AI. , issued a warning to the employer. Regarding suspicions of discrimination. The agency has not yet issued a decision on this matter. (May 2024) EEOC Investigates Allegations of AI Discrimination – In addition to filing complaints with the FTC as described above, the ACLU also unfairly singles out applicants with disabilities and filed a complaint with the EEOC regarding AON’s AI-based recruitment tools, alleging that they are targeting people. A person of a particular racial background. The agency has not yet issued a decision on this matter. (2024) First AI Lawsuit Settlement in EEOC History (sort of) – The EEOC claims that an AI-powered hiring selection tool automatically rejected female applicants over the age of 55 and male applicants over the age of 60. Settled a lawsuit against a teaching company. The company agreed to pay $365,000. to resolve charges. However, upon closer inspection, we found that the tools used by the employer were fairly rudimentary and not necessarily state-of-the-art AI systems. (August 2023) What’s next? We expect 2025 to be the year the floodgates open, with lawsuits brought against employers related to the use of AI in the hiring process and the workplace in general. We anticipate a sharp increase in lawsuits and representative actions.
global
The United States is not alone in facing the challenge of tackling AI.
EU AI Law – This landmark law was the world’s first comprehensive law regulating the use of AI when passed at the end of 2023. EU AI legislation will soon regulate the use of AI systems based on risk level, with stricter requirements in high-risk cases. AI systems – such as systems used to recruit, promote, and monitor employees. Employers must ensure compliance by conducting risk assessments, implementing transparency measures and adhering to data protection laws to avoid fines and legal liability. China – China takes an approach closer to the US than the EU when it comes to AI regulation. Recent legislation creates an AI safety governance framework, but it is not binding. Instead, it simply creates a common set of standards for AI developers to follow when developing their systems. What’s next? Countries will continue to take sides, either with strict regulatory frameworks like the EU or looser recommended principles like the US and China. We will also be focusing on Brazil in 2025. Brazilian lawmakers recently passed a strict AI safety law that establishes an unprecedented set of protections for copyrighted materials used to train GenAI systems. If finalized, the proposal could serve as a model for other countries seeking to crack down on the use of AI.
