Workers say AI is making work worse, not better

New research from Adaptavist finds 70% of U.S. knowledge workers feel nostalgic for pre-AI work and 43% would remove generative AI entirely. The findings point to a widening gap between AI adoption and employee experience, with many workers spending more time checking AI output than saving time.

Why it matters: - The findings suggest generative AI is not just changing workflows. For many workers, it is making work feel less meaningful, more repetitive, and harder to trust. - The research points to a real business risk: AI adoption can rise while employee engagement and productivity fall. - The data also shows a potential fault line between organizational rollout and day-to-day user experience.

What happened: - Adaptavist surveyed 500 U.S. knowledge workers in March 2026 as part of a broader survey of 2,500 professionals across the UK, U.S., Canada, Germany, and Spain. - 70% of U.S. respondents said they regularly feel nostalgic for how work operated before generative AI. - 33% said they preferred how work operated before GenAI tools were widely adopted. - 43% said they would remove GenAI tools from the world entirely if they could.

The details: - 19% of U.S. respondents said they had no preference either way about work before or after GenAI adoption. - Among younger workers globally, 40% of Gen Z and 40% of Millennials said they would remove GenAI from the world. - That compares with 32% of Gen X and 29% of Boomers. - 34% of workers who would remove AI said they believe it reduces creativity. - 34% cited concerns over misuse. - 38% worried about surveillance and privacy. - More than half, 54%, said dealing with low-quality AI output makes their job feel less meaningful and more repetitive. - 43% said low-quality AI output has made them less engaged at work overall. - Half of workers said they now spend more time verifying and fact-checking AI output than the time they save by using it. - 60% said poor-quality AI-generated work is slowing down their projects. - 65% said AI-generated work is reducing overall team efficiency. - 65% said their performance is now being compared, fairly or not, to AI-generated output. - 31% said they use AI simply to meet workload demands. - 30% said they use AI to keep up with colleagues. - 33% said they feel pressure to improve performance to match a machine-accelerated environment. - 35% said they feel pressure to improve quality. - 29% said they feel pressure to be more efficient. - 76% of U.S. workers said they would like their organization to increase AI use. - 79% said they trust AI is being used ethically. - 78% said their organization has been transparent about AI adoption. - 39% said they often do not understand why they are expected to use AI in their role. - 39% reported AI fatigue. - The research was conducted by Attest.

Between the lines: - The numbers suggest worker frustration is less about AI itself and more about how companies are implementing it. - The survey shows a disconnect between leadership metrics such as adoption and worker realities such as verification burden, morale, and clarity of purpose. - Neal Riley, AI Innovation Lead at The Adaptavist Group, said organizations often focus on who is using AI and how often instead of measuring its impact on the work itself. - Riley said thoughtful implementation, guardrails, and support can help AI enhance rather than erode meaningful work.

What's next: - Adaptavist’s findings point to pressure on employers to define when AI should be used, why it matters, and how its output will be checked. - Companies that want AI adoption to stick may need better guardrails, clearer expectations, and a stronger link between tools and outcomes. - The research suggests employee trust and usability will matter as much as rollout speed.

The bottom line: - Workers are not rejecting AI in full. They are rejecting the extra work, weaker quality, and loss of meaning that can come with bad implementation.