Last month, ActivTrak’s Productivity Lab released findings from one of the largest workplace behavior datasets ever assembled. Researchers looked at 443 million hours of digital work activity across 1,111 companies and 163,638 employees, spanning three years. They found that AI isn’t really reducing workloads, at least not yet. Apparently, it is amplifying them.
‘AI measurement gap’
ActivTrak found that workdays shrank by a modest 2%. But inside that compressed time, collaboration surged 34%, multitasking rose 12% and weekend work jumped more than 40%. Focus time fell to a three-year low. The average focused session now runs just 13 minutes and 7 seconds, down 9% since 2023.
At the same moment, 80% of employees now use AI tools, up 52% in two years. ActivTrak shows that the average organization runs seven or more AI tools, up from two in 2023. Time spent in AI tools increased eightfold.
“AI adoption is accelerating faster than most organizations can measure its impact,” said Gabriela Mauch, chief customer officer and head of the ActivTrak Productivity Lab, in a release.
She points out that the data shows that AI isn’t reducing work, but increasing the speed and density of how work happens. “The challenge leaders now face is closing the AI measurement gap and gaining real visibility into how AI is changing productivity, focus and workforce capacity.”
The “AI measurement gap” may be a defining HR challenge of 2026. Organizations are deploying AI aggressively and barely measuring it. According to ActivTrak’s companion survey data, 71% of respondents say they actively use or are piloting AI tools across teams. Half of them do not measure AI’s impact on their workforce.
Read more | AI skills: When employee adoption outpaces company support
What is amplified work?
ActivTrak found that employees who spend 7% to 10% of their total work hours in AI tools show the highest productivity of any usage tier. Yet only 3% of AI users fall within that range. The vast majority, 57%, spend less than 1% of their work time in AI tools. The sweet spot exists, according the report, but almost no one is in it.
After AI adoption, time spent across all measured work categories increased between 27% and a whopping 346%. Email volume rose 104%. Chat and messaging surged 145%. Usage of business management tools climbed 94%. AI users’ average daily focused time, meanwhile, declined by 23 minutes.
The report introduces a concept it calls “amplified work,” which researchers say is a state in which AI accelerates throughput without reducing the underlying cognitive load on workers. The research found that productive sessions grew 13% in length, but focus sessions shrank.
Another report, Omnissa State of Digital Workspace 2026, found that it takes an average of 23 minutes and 15 seconds to refocus after a disruption. When those disruptions happen daily, the cumulative effect is what Omnissa calls a “forced interruption tax.” Applied to an entire workforce, that tax becomes a massive, invisible productivity drain that never appears in an AI ROI calculation.
Shadow risk is the hidden variable
The amplified workplace brings with it a growth of tools employees are adopting outside of the view of IT. The Omnissa report, an analysis of telemetry data from millions of enterprise endpoints across 17 industries, documents what it calls a “shadow perimeter.”
GenAI-powered AI assistants are the fastest-growing app category in enterprise environments, notching close to 1,000% growth in 2025. IT departments are deploying sanctioned tools, primarily Microsoft Copilot, which holds a 90%-plus share of the sanctioned AI category. Meanwhile, employees are aggressively adopting publicly available tools outside IT’s view.
The split in sanctioned versus unsanctioned AI adoption may indicate a gap in user satisfaction. Employees may see corporate-sanctioned tools as insufficient or limiting. So, they find workarounds.
Personal communication tools show the same pattern. Enterprise-standard apps like Microsoft Teams, Zoom and Slack dominate the sanctioned landscape, but a majority of communication tools used on managed corporate devices are consumer apps.
Business decisions are happening on encrypted, unmonitored platforms. In regulated industries like financial services and healthcare, that is a compliance failure waiting to happen. Employees are signaling, through their tool choices, that the official technology stack is not meeting their needs.
What HR leaders can do
The convergence of these reports points toward a set of practical actions for HR leaders who are navigating the amplified workplace:
Measure AI’s impact on work design, not just output
Productivity gains are visible. Focus degradation, multitasking escalation and weekend work creep are not. HR leaders need workforce analytics that capture behavioral patterns before the cumulative cost becomes a retention or burnout problem.
Treat unsanctioned AI adoption as a workforce satisfaction indicator
When employees bypass corporate tools, they are expressing a preference. HR leaders who partner with IT to understand why employees choose unofficial tools will be better positioned to advocate for better solutions and governance policies.
Define what “healthy AI usage” looks like for your organization
The 7% to 10% productivity sweet spot identified by ActivTrak is a starting benchmark, not a universal standard. HR leaders should work with their analytics teams to establish baselines for their own industries and roles.
Quantify the cost of digital friction
The Omnissa data shows that Windows devices experience 7.5 times more “unresponsive software states” than Mac devices, and that each forced shutdown can impose a 20-30-minute refocus penalty on the employee. HR leaders who can translate these technical metrics into lost hours and dollar costs will have a far stronger argument for investment in both workplace technology and employee experience.
The post Why worker focus time is at three-year low appeared first on HR Executive.
This article was originally published on HR Executive. Click below to read the complete article.