Workplace Trends

The quiet majority now works alongside AI.

New workplace survey data shows that using an AI tool on the job has quietly gone from a novelty to a norm around the world — and the gap between people who use it a little and people who use it constantly is where the real story is.

Field Notes Desk 6 min read Workplace Survey Data
52%
of employees surveyed now say they use AI at work at least a few times a year — the first time that figure has crossed the halfway mark.

Two years ago, roughly a quarter of surveyed workers had touched an AI tool on the job. Today, more than half have — and the shift happened faster than almost anyone predicted. It's no longer a question of whether AI shows up in a workday. It's a question of how often, and for what.

The change shows up clearly in year-over-year survey data: usage has essentially doubled in two years, moving in steady steps rather than one sudden jump. That pattern suggests something more durable than a passing trend — it looks like a tool quietly becoming part of the default toolkit, the way spreadsheets or email once did.

Share of employees using AI at work, by survey wave
2 yrs ago
27%
Prior qtr
41%
Last qtr
47%
Today
52%
Figures track both individual employee usage and organization-level AI adoption reported in recent workplace polling.

01 What people actually use it for

The image of AI at work as some futuristic assistant handling entire projects doesn't quite match what's happening day to day. The most common uses are far more ordinary: getting words onto a page, tracking down information, and getting unstuck on a problem.

51%
Writing and editing
49%
Search and research
39%
General problem-solving
15%
Use it daily

02 Frequency changes everything

Employee using an AI chat tool on a laptop at their desk

About a third of employees now use AI a few times a week or more, and the survey found a clear pattern: the more tasks someone hands to an AI tool, the more positively they rate its effect on their own output. Occasional users notice a mild boost. Heavy users report something closer to a transformation.

Among employees who use AI for just one task, under half reported a clear productivity gain — but among those using it across seven or more tasks, nearly all of them did. — Workplace AI usage survey, 2026

That gap is worth sitting with. It suggests the payoff from AI at work isn't really about which tool someone picks — it's about how deeply they've woven it into their actual workflow.

03 Organizations are catching up, slowly

Nearly half of respondents said their employer has formally rolled out AI tools to boost productivity or quality, up several points from the previous quarter. But one in five employees said they simply don't know whether their organization has adopted AI at all — a sign that, in a lot of workplaces, adoption is still happening informally, tool by tool, rather than as a top-down policy.

Modern open-plan office where AI tools are part of daily work
AI adoption is spreading fastest in ordinary, everyday work — not just specialist roles.

04 Trust hasn't kept pace with use

Here's the tension underneath all this growth: people are using AI more, but they don't trust the institutions around it much at all. About two-thirds of respondents in one recent poll had little to no confidence that regulators anywhere will govern AI effectively — a number that's climbed since the same question was asked a couple of years back. Confidence in companies to build and deploy these tools responsibly isn't much higher, with roughly six in ten adults skeptical.

In other words, AI is being adopted faster than it's being trusted. Workers are folding it into their daily routines while still holding real reservations about who's steering it — a combination that tends to define new technology at exactly this stage of its life.

05 What this means going forward

If the current pace holds, AI usage at work stops being a story about early adopters and becomes a story about baseline expectations — the same way basic computer literacy or comfort with email eventually did. The workers and teams pulling ahead right now aren't necessarily the ones with access to the fanciest tools. They're the ones finding more places, more often, to actually use them.