Most discussions about artificial intelligence in the workplace assume the conflict will be technological.
What jobs will AI replace? What tasks should AI perform? How much efficiency can organizations gain?
A different question may arrive sooner than many employers expect.
What happens when an employee asks for a religious accommodation related to AI?
At first glance, the question sounds unlikely. Then again, so did many workplace issues before they arrived.
Last month, Pope Leo XIV issued Magnifica Humanitas, an encyclical addressing artificial intelligence and its implications for human dignity, work, and society. Contrary to some headlines, the document is not a rejection of technology. The concern is not that AI exists. The concern is how human beings choose to use it.
For Catholics familiar with the Church’s social teaching, the document raises real questions. Does a particular use of AI diminish the value of human work? Does it replace judgment that should remain human? Does it treat people as data points rather than persons? Does it encourage efficiency at the expense of dignity?
Reasonable people can disagree about the answers.
The more immediate challenge for employers is that some employees may eventually conclude those answers have implications for how they perform their jobs.
Imagine an employee who says:
“I am willing to do the work. I am willing to meet the expectations of the role. But I have a religious objection to using AI in the manner currently required.”
At that point, the discussion stops being theoretical, it becomes an accommodation request.
Why This Request Is Different
Most religious accommodation requests have a clear edge.
- Don’t schedule someone on Saturday.
- Permit a particular garment.
- Excuse someone from a specific procedure.
The point of friction is identifiable: The accommodation usually involves adjusting one aspect of the job while leaving the rest intact.
An AI objection may not have that shape.
If an employee objects to using AI “in the manner currently required,” the first challenge is determining what that means.
- Are they objecting to drafting assistance?
- Decision-support tools?
- Automated candidate screening?
- A triage system that routes work before a human being ever sees it?
Or are they objecting to a workflow in which AI is woven throughout the process rather than attached to any single task?
That distinction matters because many workplace technologies can be isolated and, increasingly, AI cannot.
In some organizations, AI is not simply a tool employees use. It is becoming part of how work itself is performed.
An AI accommodation request may be the first time an organization is forced to explain where the work ends and the technology begins.
This is not a case where existing precedent offers much guidance.
No court has yet addressed whether an objection to AI use qualifies for religious accommodation, what counts as a sincerely held belief in this context, or how undue hardship should be assessed when the accommodation cannot be cleanly separated from the job itself. In 2023, “undue hardship” in religious accommodation cases was defined as “substantial increased costs in relation to the conduct of [the company’s] particular business.”
Employers should not expect this to resolve into a simple yes or no, even with good legal counsel. The honest answer, for now, is that nobody fully knows.
What the Request Actually Tests
Most accommodation requests ask whether the organization can adapt to an employee’s needs.
An AI accommodation request may force a different question:
Does the organization understand its own workflow well enough to know what can be adapted?
Many organizations are deploying AI because it appears useful. Others because competitors are doing so. Some because leadership believes they should.
Far fewer have articulated where AI adds value, where human judgment remains essential, and where the line between the two should sit.
That ambiguity is exactly what an accommodation request exposes.
To determine whether an accommodation is possible, the organization must understand the work itself.
- Can this role be performed without AI?
- Which tasks depend on AI?
- Which merely benefit from it?
- What decisions require human judgment regardless of technology?
If those questions cannot be answered, the accommodation request becomes difficult not because the employee’s concern is complicated, but because the organization has never clearly defined its own operating model.
This is not an isolated gap. Recent workforce research has found that organizations are already struggling to anticipate the downstream effects of AI adoption on cost, structure, and roles, with many of those effects going unplanned simply because no one mapped them in advance. An accommodation request is one more place where that same gap shows up. The organization is asked a question about its own operations that it has not yet asked itself.
The Inside Advantage Lens
How an organization responds to a request like this depends less on its AI strategy than on whether its underlying systems are strong or rigid. A rigid system has a rule: AI is part of the job. End of discussion.
A strong system, though, has a process.
It can evaluate whether a particular use of AI is essential, whether alternatives exist, and whether the work can be accomplished differently without undermining legitimate business objectives.
That capability rests on four layers.
- Leadership practices determine whether the request is treated as a problem to eliminate or a question worth examining. Have leaders thoughtfully defined the role of AI, or has adoption simply spread because the technology was available?
- Core systems determine whether anyone has mapped how work actually gets done. Can the organization distinguish between tasks that require AI and tasks that merely use it?
- Culture and norms determine whether employees feel safe raising concerns before they become formal requests. Is thoughtful questioning encouraged, or is adoption assumed?
- Governance and accountability determine who decides where the boundary between human judgment and automation belongs. Has that authority been exercised, or is the boundary being established by default?
If those four layers have never been tested against this question, an accommodation request may become the first real test.
What This Reveals
The companies that handle these conversations well will not necessarily be the ones with the most sophisticated AI systems, or the ones with the most airtight legal answers.
They will be the ones that already understand the role they expect human beings to play in their operations well enough to reason through a question for which no precedent yet exists.
For organizations that have not done that work, an AI accommodation request will feel larger than a compliance issue. It will feel like a question the organization has never had to answer out loud, asked by someone who needs an answer now, with precedent to point to and no template to follow.
That is the real signal.
Not whether the technology works.
Not whether it increases productivity.
But whether the organization has built the kind of judgment that holds up when the answer isn’t written down anywhere yet.
The accommodation request is simply the moment that question arrives.
The Inside Advantage 
Leave a comment