When companies crunch all of your personal data when shopping or when at work and then run your data through AI-automated systems, the results are inimical not only to your personal privacy but also to your personal well-being — and to society and the economy more broadly. That’s why companies using these data surveillance systems are increasingly drawing the attention of state legislators, even as the Trump administration moves to pre-empt these efforts.
Lawmakers in Colorado, Maryland, Minnesota and New York, among other states, are moving to do more to protect workers and consumers from these harms. In Annapolis, for example, proposed legislation would prohibit certain surveillance-based price-setting as an unfair, abusive or deceptive trade practice under the Maryland Consumer Protection Act and would prohibit an employer from engaging in certain surveillance-based wage setting under state labor and employment laws. In Denver, Colorado’s House of Representatives just passed similar legislation to protect consumers and workers.
These moves are particularly timely. My recent research with co-author Wilneida Negrón examines specifically how algorithmic wage discrimination is quietly spreading from the earliest adopters, such as ride-hailing and food delivery companies, to include traditional employers in industries such as health care, customer service, logistics and retail. Our findings from a first-of-its-kind audit of 500 AI vendors that provide these services demonstrate that at least 20 of them are at high risk of enabling surveillance-based wages, and 16 of the 20 have linked their products directly into payroll or HR systems.
Under these types of surveillance wage systems, different people may be paid different wages for largely the same work, and individual workers cannot predict their incomes over time. What’s more, these pay practices — especially those that rely on so-called panopticon worker surveillance systems that provide a full view of workers on the job, alongside algorithmic intelligence or machine learning systems — result in ever-shifting wages that lead to people working longer hours but getting paid less per hour.
Many of these AI vendors boast major corporate customers, many of them operating in multiple states. The vendors of these AI surveillance systems, however, typically mask the underlying purpose of their products and services by marketing them as high-tech workforce optimization or performance management systems. Overall, however, these cross-sector tools and platforms create a troubling new standard of workforce surveillance, with plug-and-play solutions for performance monitoring, decision-making and compensation management. Without proper legal guardrails in place, workers across the country can be harmed.
What’s more, these workplace surveillance systems pose a threat to states’ collection of payroll taxes and the state programs that rely on that revenue, not least states’ unemployment insurance funds. If wages are increasingly classified as “bonuses” or some other type of non-payroll-related performance adjustment, this will render payroll tax-derived state revenues smaller and more volatile. The National Employment Law Project estimated recently that these systems are costing one state, Connecticut, $60 million a year in unemployment insurance contributions.
Unions in many of these states are now seeking to protect wage earners (and by extension, state revenue collection) against automated management and surveillance practices, but legislative action is paramount.
Veena Dubal is a professor of law at the University of California, Irvine, whose work focuses on issues of labor and inequality.
Governing's opinion columns reflect the views of their authors and not necessarily those of Governing's editors or management.
Related Articles
