On 6 May 2026, Littler Mendelson, the world's largest employment and labour law practice representing management, released its 14th Annual Employer Survey. The survey draws on responses from more than 300 US-based C-suite executives, in-house lawyers, and human resources professionals. Its central finding was striking: 84% of employers now expect business impacts from AI-related policy or regulatory changes over the next twelve months, double the 42% who said the same in 2025.
The Governance Gap Is Widening
More employers have AI policies than a year ago. The survey found that 68% of respondents now have a formal policy governing AI use in the workplace, up from 38% in 2025. That looks like progress.
The underlying picture is more complicated. Only 55% have a formal review or approval process for AI tools. Only 54% restrict the information that can be entered into them. At the same time, 37% of respondents say their organisations are reassessing job responsibilities in light of expected AI efficiency gains, while 20% report reducing or planning to reduce hiring, and 15% say the same about workforce reductions.
The gap between those numbers is instructive. Organisations are changing roles, adjusting headcount, and deploying AI tools across functions, yet fewer than half have the governance structures in place to manage how those tools are actually used or what they are actually doing to the shape of work.
Role Clarity Is Under Pressure
When firms are reassessing job responsibilities without coherent governance frameworks, the people inside those roles experience the gap directly.
The Littler survey captures a workplace in which the boundaries of many professional roles are in motion. AI is accelerating throughput across a wide range of functions, which means the question of what a person is actually responsible for, what requires their judgement, and what has been absorbed by a tool, is genuinely unclear for a growing share of the workforce.
This is not a niche problem confined to technology companies. The Littler survey covered respondents across industries and firm sizes. The pattern of AI adoption outpacing AI governance is broadly distributed.
The survey also reports that 79% of respondents expressed concern about AI-related litigation over the next twelve months, with data privacy (49%), discrimination or bias (45%), and compliance with state and local AI laws (43%) among the leading areas of focus. As Niloy Ray, co-chair of Littler's AI and Technology Practice Group, noted in the release: "AI policies should reflect how the tools are actually used by their workforces."
Most organisations have not yet closed that gap between stated policy and lived working experience.
The Accommodation Signal
Alongside the AI governance findings, the Littler survey recorded a separate but related development: 67% of employers reported an increase in mental health-related leaves of absence and accommodation requests over the past year, continuing what the survey describes as a multi-year trend. Managing extended or open-ended absences was cited as a challenge by 75% of respondents, while 70% flagged inadequate manager training as a difficulty.
The survey does not draw an explicit connection between role ambiguity and mental health accommodation trends. But the parallel trajectory across multiple years is difficult to dismiss without examination.
What does it mean for a professional to be expected to perform well in a role whose boundaries are being redrawn by decisions made above their sight line?
Opinion: The Cognitive Cost of Unclear Work
The Littler findings confirm something that has been accumulating quietly for two years. AI deployment is not simply replacing discrete tasks. It is restructuring the cognitive landscape of professional roles without providing workers with a corresponding map.
When a role is redefined through governance decisions the individual has no visibility into, the human experience of that role shifts in ways that are difficult to name. Work becomes harder to evaluate, both for the organisation and for the person doing it. Effort and outcome begin to decouple. The professional who was previously confident in a well-defined role can find themselves uncertain about where their contribution begins and ends.
That uncertainty is not a personal failing. It is a structural condition. And structural conditions respond to structural thinking, not to resilience workshops or wellness programmes.
The organisations best placed to navigate this period are those that treat role clarity as a performance input rather than an administrative byproduct. The AI governance frameworks that the Littler survey finds most organisations have not yet built are not merely legal protection. They are the infrastructure through which people understand what they are being asked to do and whether they are doing it well.
The more important question is whether ambiguity at this scale is something organisations can manage around, or whether it is now the primary variable shaping performance from the inside.
Sources
The Littler Annual Employer Survey 2026, Littler Mendelson, 6 May 2026
Employers Brace for AI-Driven Workplace Shifts and Rising Risk, Littler's Annual Survey Shows, Littler Mendelson press release, 6 May 2026
The contents of this article are for informational purposes only and do not constitute professional, legal, or financial advice.
