Civil Law

“Call it the fault of civilization. God isn’t compatible with machinery …”

Or so Mustapha Mond warns John the Savage in Aldous Huxley’s Brave New World, but that tension is not confined to great works of literature. As businesses embrace generative artificial intelligence (AI) in the workplace, a new trend is emerging: conscientious objectors requesting exemption from AI tools to accommodate their sincerely held moral views.

One may ask how AI intersects with religious observance, but prominent religious leaders and spiritual thinkers have been wrestling with this question since the advent of AI. Indeed, Pope Leo XIV’s first papal encyclical, Magnifica Humanitas, called on Catholics, Christians, and other readers to consider AI’s impact on humanity and described AI as “a valuable tool that requires vigilance,” citing the moral consequences of delegating decisions regarding employment, credit, and access to social services to automated systems that do not know “compassion, mercy, forgiveness, and … hope.” The Future of Life Institute, in which many prominent Islamic, Buddhist, and Hindu scholars participate, has publicly stressed the importance of individual agency, including the agency to opt out of AI-integrated systems. The calls to consider AI use through a spiritual lens are not limited to organized religion. The American Humanist Association (AHA) recently signed onto the Luxembourg Declaration on Artificial Intelligence and Human Values and exhorted secular humanists to examine the economic, environmental, anti-creative, and anti-competitive impacts of AI. In fact, the AHA has begun encouraging Humanists with objections to using AI in their professional lives to consult with their legal advisors.

The debate over AI ethics in the workplace is not merely theoretical. Although the trend has not yet wound its way to the courts, a few prominent AI-related religious accommodation requests have drawn public attention. For instance, in June 2026, Erin Maus, a Unitarian Universalist (UU) software engineer in North Carolina, requested and received a workplace AI exemption from her entertainment technology company employer after claiming that the ethical and environmental sequelae of AI use conflicted with her UU beliefs. Ms. Maus was permitted to write code by hand instead. More recently, David Durman, a Humanist research administrator employed by the University of Kentucky, applied for and was granted a workplace religious accommodation exempting him from using AI at work. He argued that generative AI “runs contrary to human wellbeing,” citing AI’s impact on the environment, elimination of jobs, and “theft of human creativity.” His request for exemption was granted after the University’s Office of Equal Opportunity found that it posed no undue burden on his department to allow him to perform his job duties without using AI.

US employers can expect similar requests to follow as religious and ethical leaders sound the alarm about AI. But unorthodox requests for accommodation should be old hat for US employers in the wake of the COVID-19 pandemic, when religious accommodation requests to be exempted from masking, mandatory vaccination, and other mitigation strategies soared. Lessons learned from that period are instructive in this context as well.

First, although Title VII requires reasonable accommodation to resolve a conflict between a workplace rule or requirement and an employee’s sincerely held religious belief, we learned during the pandemic (and this has been regularly confirmed by post-COVID-19 era decisional law) that little is required to demonstrate sincerity. An employer with an objective basis for questioning sincerity may make a limited inquiry, but just as employers were cautioned not to impose religious litmus tests when employees sought exemptions from public health measures, employers asked to exempt employees from AI use on religious grounds should hesitate before demanding proof of religious beliefs, such as the production of anti-AI religious texts or supporting statements from spiritual advisors. Likewise, employers should bear in mind that religious and spiritual beliefs may be sincere even when not linked to recognized, organized religion, as demonstrated by the swell of support within the Humanist community. That said, sincerity and religiosity are separate questions. Title VII protects moral or ethical beliefs about right and wrong that are held with the strength of traditional religious views, but it does not protect social, political, or economic philosophies or mere personal preferences. Because many AI objections rest on environmental, economic, or creative-rights concerns, employers should consider whether a particular objection is rooted in the employee’s religious or moral belief system or instead reflects a secular policy view.

Second, the Supreme Court clarified in its 2023 Groff v. DeJoy decision that an employer cannot deny religious accommodation merely because it imposes more than a de minimis cost or administrative inconvenience. Instead, an employer must show that the burden of granting an accommodation would result in substantial increased costs in relation to the conduct of its particular business. Speculative or hypothetical hardships will not suffice. For AI-use exemptions, employers should ask whether exempting an employee from AI technology would impose specific, measurable costs or operational burdens, such as objective impacts on productivity or an inability to satisfy client obligations, before denying a request for accommodation. Even then, employers should explore whether the employee objects to only some AI tools or to all AI technology, and should craft individualized accommodations whenever possible that balance business needs and religious expression.

Finally, employers are encouraged to carefully examine whether AI use is genuinely necessary for the role before concluding that it is mandatory. AI has undoubtedly streamlined many business operations, but the technology is still evolving. Employees have functioned successfully without this technology for decades, so employers should be careful to distinguish job duties that necessitate AI use from those where AI is merely preferred or efficiency-enhancing. [Case in point: I wrote this (almost entirely) without using AI.]

New technology presents new challenges, and AI is no exception. Employers should update accommodation policies and procedures to account for technology-use objections and train human resources professionals and first-line managers to recognize religious accommodation requests, without the need for magic words like “religion” or “spiritual” objection, giving due respect and consideration to employees’ bona fide objections to this morally complex emerging technology.

Home – LexBlog