Workers Skeptical of AI, Seek Fewer Apps and Value Human-Centric Skills Most
Halfway through the year is a good time to reflect on what the prevailing and emerging trends in workplace research are shaping up to be. Here’s what we can conclude so far:
Workforce members remain ambivalent about the role AI will play in their jobs
Multiple surveys have logged workers’ persistent resistance to any hype about AI improving their daily work experience: In January, Unily’s “AI Reality Check” found that only 20% of employees thought AI was a “must-have” for business competition, and “36% believe that more AI adoption would have no impact on their company’s performance.”
This past week. GoTo published its report “The Pulse of Work in 2025: Trends, Truths, and the Practicality of AI,” and clocked both the AI skepticism among workers (“62% of employees believe AI is significantly overhyped”) and the premise that AI will free up worker time to boost productivity — something the workers are far less prepared to believe (86% believe the technology is not very accurate or reliable) compared to their managers (only 53% of managers are as skeptical).
A major productivity challenge is still hampering workers
Switching between applications is an enduring drain on worker productivity and time, and reducing this friction and keeping workers in one continuous environment has long been a goal of enterprise collaboration platforms — witness Microsoft pushing for this as far back as 2018, the rise of whiteboard apps in 2021-2022 to try and keep folks centered in one collaborative space, Zoom announcing products meant to reduce mode-switching in 2023, Slack advocating for a reduction in app-switching later that year, and vendors kicking off 2024 with the news they were focusing on reduced app-switching.
One of the more recent arguments in favor of agentic AI is the promise of orchestrating agents to tap specific apps and automate workflows, so the AI bots are the ones switching through apps, not the end users. In February this year, vendors began to connect the dots to prior research:
In Asana’s Anatomy of Work Global Index 2023, the term “busy work” didn’t come up, but the work management platform did define something similar and the numbers weren’t great:
“work about work” [is the] hours spent on duplicated work, unnecessary meetings, and juggling too many apps. Work about work takes up 58% of the workday, with skilled work taking up 33% and strategic work just 9%.
In the Anatomy of Work Global Index, a major boost to productivity was identified — and it wasn’t AI. Rather, workers complain about having to use too many applications, and the app juggling makes more work around trying to do the “real” work of collaborating, acting on new information, or doing assigned work in a timely fashion:
Among workers who use more than 16 apps, 25% said they miss messages and actions, compared to 8% using 1-5 apps and 15% using 6-15.
Loss of focus is another side effect—23% of workers using 16 or more apps said their attention span was reduced because of app-switching.
The more apps people use, the more likely they are to say they are less efficient. Of those using 16 or more apps, 26% said they were less efficient because of app switching.
Will adding an AI tool help with the app-switching conundrum? It depends on whether AI will streamline existing workloads or if its output will be yet another thing to review and manage before getting on to the skill-based or strategic work. Workers don’t run toward technologies — they adopt solutions to their problems.
Savvy workers are using AI to boost their value as humans
As more workflows get automated, more employers are realizing exactly what people bring to the equation — soft skills that allow them to be present and adaptable. According to LinkedIn, seven out of 10 of the “skills on the rise” in demand on the employment platform are human-centric skills like customer engagement and conflict mitigation.
And people’s critical thinking skills may be more valued as more people use generative AI — a study from Microsoft and Carnegie Mellon that found “the more confident the worker was in the AI’s capability to complete the task, the more often they could feel themselves letting their hands off the wheel,” suggesting a slackening in the type of analytical thinking or creative generation of ideas. So leaning into the critical thinking by treating AI with skepticism may boost both the AI-driven outcome and the human performance shaping it.