The State of Work-Life Wellness 2027

The workforce is splitting by AI fluency

We surveyed 5,021 full-time employees across the globe and introduced a new framework to understand how AI is reshaping wellbeing: the AI Fluency Score. Every respondent received a score from 0 to 9 based on current AI usage, adoption confidence, and agreement with a set of fluency statements, sorting the workforce into three groups.

AI fluency is now a wellbeing fault line. Employees are no longer splitting mainly by generation or role when it comes to workplace wellbeing. How much AI someone uses, and how confident they feel using it, now predicts burnout, job security, and resilience.

The employees driving AI adoption are absorbing the most strain. The workforce most responsible for AI integration inside companies today is also the one carrying the heaviest daily burnout and the steepest workload increase, a cost currently falling on early adopters.

Feeling equipped matters more than using the tools. Resilience tracks far more closely with whether someone has been given time and support to build AI competence than with how long they've been using AI at all, which makes it a condition organizations can build.

Clear AI policy functions like a wellbeing benefit. Employees who understand their company's AI policy report meaningfully higher resilience, regardless of their own fluency level, evidence that policy clarity itself is a form of psychological support.

Wellbeing is now inseparable from AI-era retention. The employees most critical to AI-era performance are also the most likely to leave without wellbeing support, making retention strategy inseparable from wellbeing investment.

A year ago, companies were still early in their AI journey and the conversation about this new technology's impact on work was still mostly theoretical. Would it replace jobs? Would it make us faster? Nobody was quite sure what daily work with AI would actually feel like. That conversation is over. Employees now know exactly what it feels like, because they're living it: faster deadlines, higher expectations, and a heavier information load, as day-to-day work shifts from execution to judgment and quality assurance.

To understand what that shift is actually doing to people, Wellhub surveyed more than 5,000 full-time employees across 10 global markets and introduced a new lens for this year's research: the AI Fluency Score. It sorts the workforce into three groups, Beginners, Adopters, and Pioneers, based on how much they use AI and how confident they feel doing it. What we found is a workforce splitting along a line nobody was tracking. AI fluency is now a strong predictor of burnout, job security, and resilience.

The employees driving AI adoption inside their companies, the Pioneers, are absorbing the most strain from it. They report the steepest workload increases and the heaviest daily burnout of any group, often while coaching colleagues on the side, sometimes unpaid and unplanned for. And yet they also report feeling better about their wellbeing than colleagues who are less fluent with AI: physically, mentally, and socially. Two things help explain why. The first is that Pioneers do not wait for the pressure to pass. They know that being well makes them better equipped to handle what AI adoption demands of them, so they prioritize it: 41% build a wellness activity into the workday at least once a day, compared to 24% of Beginners. The second is that they are not doing it alone. Sixty percent say their manager actively backs their wellbeing, compared to 31% of Beginners. One possible reason: Pioneers may simply have better managers overall, the kind who guide them through the AI shift and help them prioritize their wellbeing at the same time.

That distinction, between support that leaders or companies in general think exists and support that an employee actually feels, runs through the rest of this year's findings. Employees at companies with a clear, well-communicated AI policy report resilience at nearly double the rate of those who aren't sure a policy exists at all: 77% versus 39%. A clear AI policy spells out what's expected of employees, not just what's technically allowed, and gets communicated and modeled until people actually know it. What separates those two groups isn't the document. It's whether anyone ever explained it and whether leaders visibly work by it, effectively role modelling the expected behaviors. A policy nobody has heard of does the same thing for an employee as no policy at all.

The same holds for wellbeing support, which is more than a line item on a benefits list: 87% of employees say they perform better when they prioritize their wellbeing. The most concerning part: Pioneers, whose skills are the hardest to replace, are also the most likely to leave if that support disappears, exactly the people a company can least afford to lose.

We measured what happens when wellbeing support is genuinely in place. Employees at organizations offering Wellhub report better outcomes than employees without it on every single wellbeing dimension we measured this year, sleep, culture, resilience, mental and physical health. The folks in charge of the benefit see the positive impact too, in lower healthcare costs, higher productivity, better retention of talent. None of it required slowing AI down, only building the infrastructure to actively support employee wellbeing underneath it at the same pace.

AI is not going to wait for organizations to catch up. But the advantage won't go to whoever adopts it fastest. It will go to whoever looks after the wellbeing of the people adopting it. This year's data reinforces that moving fast and taking care of people were never really a tradeoff. No matter the technology change or business challenge they're the same decision and the only way to ensure sustainable results.

Most workweeks now end the way they used to end only during a crunch: one more notification after dinner, one more question that "will only take a second," one more tab left open past the hour anyone meant to close it. That used to be the exception, reserved for a launch week or a closing quarter. For a growing share of the workforce, it's just Tuesday, and it's happening at the exact moment employers are being told AI is finally about to make work easier.

That gap between the promise and what people are actually living through is where this chapter starts, because the one thing that hasn't changed in any of this is the human being doing the work. People still need movement, rest, food, and contact with each other to function, the same requirements that have shaped how humans work for centuries. What has changed is the speed at which work moves around those needs. AI is compressing timelines and raising the baseline for what counts as a normal day's output faster than any previous shift in how people work, and it's doing so before most organizations have adjusted the conditions employees need to keep up. Stress, burnout, and disengagement aren't a mystery showing up in this year's engagement survey. They're the predictable output of a body being asked to run past the limits it was built to sustain.

Start with the most basic input: sleep. Adults need seven to nine hours a night for full physical and cognitive recovery. Consistent sleep at that threshold measurably improves memory and impulse control compared to insufficient sleep (National Sleep Foundation / American Academy of Sleep Medicine; Sleep Health), which means an organization running on a sleep-deprived workforce is running on people who are measurably worse, in a literal, clinical sense, at the judgment calls their jobs depend on. Treat that as a structural constraint on how decisions get made, not a line on a wellness checklist.

Connection works on the same principle, even though it rarely gets treated with the same seriousness. The physical health effects of chronic loneliness are comparable to smoking fifteen cigarettes a day, and poor social connection raises the risk of heart disease by 29% and stroke by 32% (U.S. Surgeon General, "Our Epidemic of Loneliness and Isolation"). Put next to a sleep statistic, a stroke-risk statistic reads like it belongs in a different report, one about public health rather than workplace culture. And there lies the issue. Few leaders make the connection: a 32% jump in stroke risk would trigger a safety review anywhere else in the business. Loneliness carries that same risk, unnoticed. It deserves the same response, not as a cultural nicety, but as a measurable health risk.

The long view backs this up. The Harvard Study of Adult Development, one of the longest-running studies of adult life ever conducted, found that how satisfied someone was with their relationships at age fifty predicted how healthy they'd be at eighty more reliably than their cholesterol levels, or any other midlife health marker researchers measured. Decades later, connection outperformed the biomarkers doctors actually watch for. That's a strange thing to be true, and it's also the clearest evidence available that connection isn't a soft variable sitting next to physical health. It's one of its better predictors.

None of this is a new discovery. What's new is that the current pace of work leaves so little room for either one, and that shortfall is quietly becoming expensive for organizations that still file rest and connection under personal life rather than performance.

Here's the overload employees are already living inside, before AI enters the picture at all: 85% of HR leaders say information overload at work is negatively affecting employee mental health (Wellhub, Return on Wellbeing 2026). Employees are saying something close to the same thing from their own side: 42% report their work stress has increased over the past 12 months. That number isn't an abstraction, because the structure of the workday backs it up in granular detail. Employees are interrupted every two minutes during core work hours, averaging 275 interruptions a day (Microsoft, 2025 Work Trend Index). Each of those interruptions carries a real cost to recover from: an average of 23 minutes and 15 seconds to fully regain focus after a significant one (Gloria Mark, UC Irvine). Do that arithmetic across a single day and the eight hours on the calendar stop resembling the eight hours anyone actually gets to spend doing focused work.

AI arrived with a specific promise attached to it: it would hand some of that time back. However, 30% of employees agree their company's use of AI has increased their workload rather than reduced it. Instead of subtracting work, AI adds a layer of prompting, verification, and oversight on top of the work that already existed, work that still has to fit inside the same eight or nine hours it always did. The logic of the tool and the logic of the day don't match. A tool built to speed up an individual task does nothing to protect the sustained attention that task still requires, so a faster task doesn't automatically add up to a lighter day, and in a workday already fragmented by hundreds of interruptions, it may not add up to anything at all. For an HR leader who rolled out AI tools on the promise of time saved, here's the test: check that promise against what employees actually report. Don't assume the pitch and the outcome are the same thing.

The other half of the AI story is happening in headcount, not hours. Forty-one percent of employers anticipate reducing staff as AI automates tasks previously done by humans (World Economic Forum, The Future of Jobs Report 2025), and this isn't a forward looking guess so much as a trend already underway. Global tech layoffs reached nearly 154,000 in the first half of 2026 alone, on pace to exceed the 246,000 recorded across all of 2025, with AI cited as the leading driver (Layoffs.fyi, via Gulf News).

It's worth being precise about what a layoff actually does inside a company that's simultaneously rolling out AI. It doesn't just remove a job. It takes the output that a job used to produce and redistributes it across the people still there, at the exact moment AI is raising the bar for what "output" is supposed to look like in the first place. Whoever remains on a team after a reduction is now doing more, faster, with fewer people to share the load. That's two forces compounding at once, on the same person, and it's exactly the kind of combination that tends to show up first in burnout data, well before it shows up as a resignation.

Layer one more shift on top of the first two: the boundaries that used to mark the start and end of a workday are eroding. After-hours chats are up 15% year over year, with employees sending an average of 58 messages before or after standard working hours. Meetings scheduled after 8 p.m. are up 16% year over year, and 30% of meetings now span multiple time zones, up eight points since 2021 (Microsoft, 2025 Work Trend Index).

Any one of those numbers on its own looks like a minor scheduling quirk. Together, they describe something bigger: a workday that has stopped ending at a fixed point and has started quietly expanding into hours that used to be off-limits by default, not by policy but by habit. Those are the same hours a workforce already short on sleep and connection, per the research earlier in this chapter, would otherwise be using to rebuild both. Work isn't just getting longer. It's colonizing the exact recovery time people need most, at the exact moment they need it most.

Put the three shifts together and a pattern emerges that's bigger than any one of them. Compressed timelines, leaner teams, and a workday without edges are already defining how workplace wellbeing conversations are unfolding in 2026, even as the technology driving all three is marketed as a source of relief (Global Wellness Institute, "Workplace Wellbeing Initiative Trends for 2026"). For a growing share of employees, this isn't a temporary spike tied to one bad quarter. It's the new rhythm of the job itself, built on acceleration, uncertainty, and sustained pressure with no obvious end date.

That pressure doesn't land the same way on everyone, though, and this is where the story turns from something universal into something an HR leader can actually act on. How much of this strain any individual employee is absorbing right now depends heavily on where they sit in relation to AI itself. Twenty-six percent of our survey respondents are still AI Beginners. Thirty-two percent are Adopters, further along but still building confidence. Forty-one percent are Pioneers, using AI as a core part of how they work. That split predicts more about how much strain someone is carrying than their tenure or their role does, and it's the lens the rest of this report uses to explain why the same technology can feel like relief to one employee and an added burden to the one sitting next to them.

Ask a company what it does for employee wellbeing and the answer comes quickly: a gym benefit, a counseling line, a manager who checks in. Ask employees whether they feel supported and the answer comes slower. Both sides are being honest. The investment is real, and so is the experience of not feeling it, and the space between those two truths is where this year's most actionable findings sit.

Employees are adjusting to that space. Seventy-five percent say their employer bears some responsibility for their wellbeing, down from 81% last year and 85% the year before. They aren't deciding they need less help. They're revising their expectations to match their experience, and an employee who has stopped expecting support is a bigger retention risk than one who's still complaining about it. A frustrated employee is still negotiating. A resigned one is comparing offers.

When employees describe the support that does reach them, three sources come up again and again: a clear AI policy, a manager who backs them, and a wellbeing program they can actually use. Those three levers share a flaw. All three deliver most reliably to the employees who are already doing best.

An AI policy sounds like a technology decision, something for IT and legal to settle. The wellbeing data say it belongs on HR's desk too. How fluent employees are with AI, and how supported they feel while building that fluency, now shapes their burnout, their job security, even their sleep. A policy is where that support becomes official. It tells every employee what's allowed, what's expected, and what they'll be judged on, and a real one is a working program: expectations people can repeat from memory, communicated often, and modeled by leaders who follow the same rules themselves. A document that sits in a shared drive and appears once during onboarding is not a policy. To an employee, it's silence.

Thirty-four percent of employees say their company has no formal AI policy at all, and the silence is loudest for the people who need guidance most: 44% of Beginners report no policy, compared to 36% of Adopters and 26% of Pioneers.

Pioneers can work without stated rules because they've had time to build their own judgment. Beginners, already the most anxious group about AI, are left guessing at the exact question a policy exists to answer: is using this encouraged here, or will it be held against me?

Knowing where you stand turns out to be worth a lot. Seventy-seven percent of employees at companies with a clear, well-communicated AI policy feel resilient and able to adapt to the pace of change, compared to 39% of those who aren't sure a policy even exists. Both groups work with the same technology at the same pace. The difference is that one of them isn't spending the day wondering.

For HR, this is the fastest lever of the three, and the mistake is thinking the document is the deliverable. The 77% belong to companies where the policy is communicated and demonstrated: leaders who use the tools openly, expectations repeated until they're familiar. Companies spend heavily on AI tools and training. The data suggest the cheapest thing they could offer employees is certainty.

Everything a company intends for employee wellbeing travels through one person. The program can be generous and the policy can be clear, but it's a manager who decides, mostly in small moments, whether any of it is usable: whether stepping out for a workout reads as normal or as slacking, whether a calendar block for a walk survives the week, whether anyone ever asks how someone is actually doing. Manager backing was one of the two conditions that explained why Pioneers stay well under the heaviest load. This is where that condition gets distributed.

It is not distributed evenly. Sixty percent of Pioneers say their manager actively supports their wellbeing, not just their output, compared to 42% of Adopters and 31% of Beginners, a 29-point spread. In 1971, a Welsh physician named Julian Tudor Hart described what he called the inverse care law: good medical care tends to reach the people who need it least. Half a century later, wellbeing support at work is obeying the same law. Beginners report the worst wellbeing on every pillar and the most fear about their future, and they are the least likely to have a manager backing either. That's a loop that feeds itself. A struggling employee doesn't stop getting noticed; they stop getting backed. Less backing leads to worse wellbeing, worse wellbeing drags down performance, and declining performance draws scrutiny instead of the positive, constructive attention that helps a person recover.

Managers aren't doing this carelessly. Support has become a retention tool, and managers spend it on the people whose departure would hurt most: 82% of employees with a strongly supportive manager say they'd consider leaving if that support disappeared. Most managers aren't even making a deliberate choice. They're overloaded: 51% of organizations flag manager overload as a significant risk to company performance (Wellhub, Return on Wellbeing 2026). A stretched manager protects their top performers first. Everyone else is left to manage on their own.

What makes that triage expensive is that it forfeits the manager's real power. The same attention that keeps a Pioneer from leaving can turn a Beginner into a Pioneer, and an underperformer into a high performer. The 29-point spread isn't just a map of where support currently goes. It's a map of conversions nobody is attempting, at the exact moment companies need more fluent, resilient employees than they can hire. So the fix runs on two levels. Make baseline wellbeing support standard for every employee, part of how the company operates rather than something a manager hands out, so nobody has to be noticed to be supported. That frees the manager's limited attention for the thing only a manager can do: developing the people with the furthest to climb.

Of the three levers, the wellbeing program is the one with the most reach. A clear AI policy removes uncertainty and a supportive manager protects time, but both of those clear the way. The program is what fills it: the actual workout, the therapy session, the nutrition plan, the night of sleep. It's the only lever that touches every pillar of wellbeing at once, and the only one that reaches every employee directly instead of depending on who their manager happens to be.

Only 26% of employees rate their company's wellbeing program as good this year, down from 29% last year and 41% two years ago. That decline is not fading interest in wellbeing. Employees are exercising more than they have in three years, half of them are in therapy, and when asked what support they're missing, they ask for more in every category. Demand for what the program offers is rising while the rating of the program falls. What employees are marking down is the fit between the program they were given and the workday they actually have.

The clearest evidence is time. Forty-six percent of employees say their company does not give them the time and space in the workday to manage their wellbeing. Usage follows that permission: 41% of Pioneers build a wellness activity into the workday at least once a day, compared to 31% of Adopters and 24% of Beginners, the same groups whose managers most and least protect their time. A program can be well designed and still unusable, if using it requires an hour the day never gives.

That's the test a modern program has to pass, and passing it takes three things.

The first is accessibility. Support has to fit into the openings a real day offers: a class near the office at lunch, a gym near home in the evening, ten minutes of meditation between calls. When wellbeing can happen in the gaps the day already has, no one has to negotiate for a bigger one.

The second is variety. Ask employees which kinds of wellbeing support they need but don't have, and they name every category at nearly the same rate:

Employees aren't asking for one missing benefit. They're asking for all of it, because the pillars hold each other up: a full night of sleep makes the morning workout likelier, the workout steadies mental health, a decent meal fuels all three, and a class taken with other people rebuilds the connection the workday has been quietly draining. Support one pillar and you've helped it. Support them together and each one starts reinforcing the others.

The third is range. An employee who hasn't exercised in years needs a gentle way in: a beginner class, a walk that counts. An employee who trains five days a week needs options that keep pace with them. A single offering can't serve both, which is why programs built around one gym or one app keep disappointing someone: whoever it wasn't designed for.

One measure tells a benefits team whether a program passes all three tests. Enrollment counts who signed up. Engagement counts who keeps coming back, and only a program that's easy to reach, broad enough to matter, and built for every level keeps earning that return visit.

Step back from the chapter and the three levers tell one story. The AI policy exists, but the employees who most need the rules never heard them. The manager cares, but the caring goes to the people already thriving. The program is funded, but half the workforce was never given the time to use it. None of this was designed. It's what happens by default when a company assumes support will find its way to people on its own. The companies that close the gap will do it deliberately, from both ends at once: a program accessible, varied, and engaging enough to put a floor under every employee, and managers freed to do what only managers can, lifting the people with the furthest to climb. What that's worth, in retention and in performance, is where the data go next.

Everything to this point describes what the right wellbeing support would look like. It would fit into the openings of a real workday. It would cover every pillar instead of forcing a choice between them. It would work for the employee who has never tended to their wellbeing and the one who has trained for years. And it would reach everyone, not just the people a manager is already protecting. That reads like a wish list. The rest of this chapter is what happens at companies where it's an actual benefit: employees at organizations offering Wellhub, compared with employees at organizations without it.

This comparison holds one thing constant: every employee in it works for a company that funds a wellbeing program. The only difference is which one.

Sixty-seven percent of employees with Wellhub rate their overall wellbeing as good or thriving, compared to 51% of employees with another program. In a company of 1,000 people, that difference is 160 employees doing well instead of getting by, decided by which program the company picked.

The advantage holds on every pillar, and it's largest on the hardest ones. Employees with Wellhub rate their mental wellbeing higher, 68% compared to 51%, and their physical wellbeing higher, 69% compared to 52%. On fitness, it's 66% compared to 42%. On nutrition, 59% compared to 36%. And on sleep, the workforce's weakest pillar, 55% compared to 28%: nearly double.

The order of those results is the story. Nutrition and sleep are the pillars other programs make the least headway on, and they're exactly where Wellhub employees pull furthest ahead. That's what breadth buys: support that reaches the pillars a gym benefit alone never touches.

A program works when employees can actually use it: when it's accessible, varied, and built for every level. Those are design choices, and they're the ones Wellhub makes. One membership covers fitness, mindfulness, nutrition, and sleep, so an employee never has to choose which part of their wellbeing deserves the benefit. The same membership serves the employee taking their first class in years and the one who trains five days a week, which is how it reaches Beginners and Pioneers instead of one end of the workforce. It fits the openings a real day offers: a class near the office at lunch, a gym near home in the evening, a session in the ten minutes between calls. And because so much of it happens with other people, a class, a gym, a familiar face at the same hour, it carries the accountability that keeps a habit alive past January.

Multiply that across a workforce and it becomes something bigger than personal habit. One employee slipping out to a lunchtime class is a routine. Hundreds doing it is a norm: the class is on calendars, people talk about going, and colleagues join instead of raising an eyebrow. And a single shared membership is what lets managers lead it. A manager on Wellhub holds the same benefit as their whole team, so when they take the noon class or the walk between calls, they're not exercising an executive perk, they're visibly using the same thing they're encouraging everyone else to use. Every open use is permission the team can see, and no memo grants permission as convincingly as a manager's calendar. Employees can tell when that's happening. Seventy-one percent at Wellhub companies say wellness is genuinely built into how their company operates, compared to 32% of employees with another program: the widest difference on any measure this year. Resilience follows: 76% feel able to adapt to the pace of change at work, compared to 57%, and 70% say they're happy at work, compared to 43%. A workforce that recovers a little every day has more left when the pace picks up.

Employers don't experience their people's sleep or nutrition directly. They experience what those things produce: the healthcare claims, the output, the way a team handles a hard quarter. On every one of those, organizations offering Wellhub report better results than organizations running other wellbeing programs (Wellhub, Return on Wellbeing 2026).

Start with the number a CFO checks first. Seventy-three percent of organizations using Wellhub report reduced healthcare costs, compared to 45% of those using other programs. Wellbeing that goes unaddressed long enough becomes a medical claim, which makes healthcare spend the hardest-nosed measure a wellbeing program can be judged on, and it's where the difference is largest.

The performance measures follow the same order. Ninety-five percent of HR leaders at organizations with Wellhub report improved employee productivity, compared to 88% at organizations with other programs: most wellbeing investment helps productivity, and this one helps more often. Seventy-five percent of HR leaders at Wellhub organizations report improved employee mental health, compared to 59% at organizations with other programs. And 90% say Wellhub has made their employees more resilient in the face of work-related challenges, compared to 82% of HR leaders crediting other programs with the same.

The company funds the benefit. The employee uses it. Both sides report that it's working, and they're describing the same chain from opposite ends: employees perform when their wellbeing holds, wellbeing holds when support actually reaches them, and support reaches them when it's built the way people are: accessible, varied, and shared. That's the Wellhub effect.

The tools of work have changed more in three years than in the thirty before them. The people using those tools haven't changed at all. They still need sleep, food, movement, and one another, and they need those things most during exactly the kind of upheaval AI has brought to their jobs. Any company's AI strategy is, whether it says so or not, a bet on how long its people can sustain the pace of change. That bet is winnable. It has conditions.

The employees keeping pace with AI are not tougher than everyone else. They're better supported. They make time for their wellbeing inside the workday, and a manager protects that time. They know what their company expects of them with AI, so they aren't carrying uncertainty on top of the workload. They feel secure enough about their future to sleep at night. Take those conditions away and the same person struggles; put them in place and a struggling person catches up. Nothing found this year suggests a resilient kind of employee exists. There are only supported employees and unsupported ones.

For an employer, that reduces to three moves. The first is a clear AI policy: write it, communicate it constantly, and have leaders visibly work by it, because certainty is the cheapest wellbeing a company can provide. The second is empowered managers: encourage them to actively back their people's wellbeing, not just their output, to protect the time that wellbeing takes, and to model it in the open, the noon class on the calendar, the walk between calls, so that an employee's support stops depending on which manager they happen to have. The third is one wellbeing program with real reach: spanning the pillars, meeting every level, fitting inside a real day, because the pillars rise and fall together, and a program that covers two of them was never going to hold up the rest.

The return on those moves is not a matter of sentiment. Employees say almost unanimously that they perform better when their wellbeing holds. The companies whose employees report the highest wellbeing outperform their peers on profits and market returns. And the people most likely to leave when support is missing are the ones with the scarcest skills and the most recruiters in their inbox, in departures half of them say their company could have prevented it. An AI investment pays off through people or it doesn't pay off at all.

Wellhub is built for exactly this. One membership spans fitness, mindfulness, nutrition, and sleep, with the range to serve an employee taking their first class in years and one who trains five days a week, in whatever opening the day offers. Because the whole team holds the same benefit, a manager using it grants permission just by being seen, and wellness moves from a benefit some individuals use to how the company operates. Employees at organizations with Wellhub report better outcomes on every pillar than employees with other programs. Their employers report lower healthcare costs and stronger productivity. Both sides are describing the same thing: support that reaches the people it was bought for.

The companies that come out ahead of this shift won't be the ones that adopted AI the fastest. They'll be the ones that supported the people adopting it.

Wellhub conducted its State of Work-Life Wellness 2027 survey to assess the current status of worker wellness and examine how well employee benefits packages are meeting worker needs.

Between June 12, 2026 and June 19, 2026, more than 5,000 full-time employees were surveyed online through the polling agency QuestionPro. The response pool included only workers 18 and older who have worked a full-time job in the last three months. The results have a 95% confidence level and a 5% margin of error.

Response options for the survey’s 86 questions included Likert scales, multiple-choice, and multi-select options.

The countries included in this survey were: United States, United Kingdom, Brazil, Argentina, Chile, Spain, Italy, Germany, Mexico, and Romania. Between 500 and 505 responses were collected from each country. Where necessary, question sets and response options were adapted to reflect regional variations in benefit availability.

Respondents were sorted into three AI segments, Pioneers, Adopters, and Beginners, using a combined AI Fluency Score built from three inputs. The first captured how respondents currently use AI tools such as ChatGPT, Copilot, Claude, and Gemini, on a scale from never having used one to using AI daily and actively coaching colleagues on it, scored zero to three as stated. The second captured respondents' own AI anxiety, from feeling overwhelmed or distrustful to feeling fully comfortable and trusting of AI, also scored zero to three, with lower anxiety producing a higher score. The third asked respondents to agree or disagree with nine statements about their day-to-day use of and reliance on AI at work, and scored them zero to three points based on how many they agreed with: zero agreements scored zero points, one or two agreements scored one point, three to five scored two points, and six to nine scored three points.

The three scores combine into a total ranging from zero to nine. Respondents scoring seven to nine are classified as AI Pioneers, four to six as AI Adopters, and zero to three as AI Beginners. Every wellbeing comparison broken out by these three groups throughout the report uses this score as its basis.

Perspective commentary throughout this report is provided by members of the Wellhub US Advisory Board, who are compensated by Wellhub for their advisory services.