How is AI changing work? It’s the question on every leader’s mind. But much of what you’ve heard about AI in the workplace is incomplete. In some cases, the predictions miss the mark.
Leaders expected AI to be deeply embedded in daily workflows. Employees thought it would replace a large chunk of work. And everyone, it seems, assumed AI tools would rapidly transform how work gets done.
But that was before ActivTrak released the fifth annual State of the Workplace report. Now, for the first time, we have objective data on how AI actually changes work.
It’s not what you’d expect.
We analyzed behavioral work data across 1k+ companies to spot real AI trends.
For the 2026 State of the Workplace report, ActivTrak’s Productivity Lab examined daily work activity across 163,638 employees and 1,111 companies spanning industries like finance, healthcare, insurance and technology.
It’s one of the most comprehensive and objective studies of work behaviors to date, including more than 443 million hours of actual daily activity. Rather than surveys and self-reported sentiment, the team observed how work actually unfolds day to day — and the role AI plays.
Here’s what the data says about AI adoption, usage and productivity.
To help leaders understand what’s really changing, we analyzed how real-world data stacks up against widely cited AI predictions. This comparison reveals the gap between what many organizations expect from AI and how it actually impacts day-to-day work.
1. AI tool sprawl is currently the norm.
AI adoption is rising quickly, but it’s happening in a fragmented and highly experimental way.
The prediction: AI will be deeply embedded in workflows in 2026.
Microsoft, April 2025: “81% [of leaders] say they expect agents to be moderately or extensively integrated into their company’s AI strategy in the next 12–18 months…24% of leaders say their companies have already deployed AI organization-wide, while just 12% remain in pilot mode.”
Gartner, August 2025: “40% of enterprise applications will be integrated with task-specific AI agents by the end of 2026, up from less than 5% today.”
ActivTrak data: Most organizations are still in experimentation mode.
The number of employees using AI jumped from 52% in 2023 to 80% in 2025. But the response is not consolidation or deep integrations with a few trusted platforms. Instead, organizations are expanding across multiple tools — likely because they’re still in the early stages of exploration. The average organization now uses 7 AI tools in 2025, up from 2 in 2023. And 83% use 6 or more AI tools.
What this means for your organization: It’s time to move from experimentation to strategy.
Regardless of your organization’s maturity level, employees are using AI. Which means it’s time to govern — not just greenlight — the tools they use. If you don’t yet have an AI workplace policy in place, start building one today.
Not sure where to start? Consider creating guidelines for ChatGPT use. Employees use ChatGPT more than any other AI tool — 27x more than the next most-used AI assistant, Cursor — making it a good starting point for AI governance.
2. Most employees barely use or save time with AI.
Despite high adoption rates, meaningful use (and measurable time savings) remain low.
The prediction: AI will replace large chunks of work and free up hours of time for high-value tasks.
McKinsey & Company, January 2025: “[Employees] believe that AI will replace 30% of their work in the next year.”
Thomson Reuters, June 2025: Professionals across multiple knowledge sectors expect AI to free up 5 hours a week, or nearly 240 hours per year.
ActivTrak data: Among employees who use AI, most spend less than 1% of their work hours in AI tools.
The largest segment of AI users (57%) spend less than 1% of their total hours in AI tools — and the “sweet spot” is nowhere near the 30% prediction in McKinsey’s survey.
In reality, the most productive AI users spend 7–10% of their work days using AI. But to date, only 3% fall within that range.
Meanwhile, data shows AI is amplifying work — not replacing it. When comparing activity 180 days before and after AI adoption, a subset of 10,584 employees spent more time on other activities after using AI. And not just incrementally. Email activity went up 104% and chat and messaging increased 145%.
What this means for your organization: It’s important to approach AI as a productivity enhancer, not a substitute for existing work.
High-performing employees clearly lead the way. But rather than using AI to reduce effort, they’re increasing output. AI helps them move faster, take on more work and expand their impact across tasks.
The key is to define what “good” looks like. That starts with setting realistic productivity baselines, identifying where AI meaningfully improves output and aligning teams around practical use cases that drive measurable results.
3. Few leaders can measure AI impact.
AI adoption is accelerating, but measurement practices have not kept pace.
The reports: Few organizations are measuring the ROI of AI in a disciplined way.
Thomson Reuters, February 2026: “82% of [leaders] say their organizations are either not collecting ROI metrics around AI usage or they are unsure about whether their organizations are collecting such metrics.”
KPMG, September 2025: “78% [of leaders] agree or strongly agree that traditional business metrics are becoming insufficient in measuring AI’s impact.”
ActivTrak data: There’s a fast-growing need for better AI adoption and impact measurement.
Various studies, including ActivTrak’s research, point to the same conclusion. Most organizations face a big gap between AI adoption and understanding its real impact on productivity, focus and workforce capacity.
In a recent ActivTrak customer survey, 50% of leaders say they don’t yet measure AI impact. Yet one-third are concerned about AI security and data privacy, and another 47% worry about ensuring effective AI use or struggle to differentiate between actual productivity gains versus hype. (Which is why ActivTrak will soon roll out special AI insights features — more on this below.)
What this means for your organization: You need to understand how AI is changing work.
The companies seeing results aren’t just adopting AI — they’re measuring impact. They understand not only which tools teams use, but which ones drive measurable productivity gains.
Benchmark and improve AI adoption with ActivTrak
AI is already changing how work gets done. But the impact is uneven and, in many cases, not yet fully understood. Organizations that take a more disciplined approach to adoption and measurement will be better positioned to turn early experimentation into sustained performance gains.
ActivTrak is here to help. Our workforce analytics platform provides a clear view of how employees use AI and where it drives results. And in mid-2026, we’ll roll out the all-new Adoption & Impact dashboard. This upcoming release will allow leaders to assess AI adoption across teams, see where AI usage drives real ROI and decide where to invest next.
Ready to get started? Sign up for a free account to start collecting and analyzing your workforce data today. Or get in touch to learn how ActivTrak can help you turn AI adoption into a competitive advantage.
