According to the latest KPMG Q3 2025 AI Quarterly Pulse Survey, adoption of AI agents in major U.S. organizations is entering a period of hypergrowth, and workplace culture and management strategies are evolving rapidly as well. A survey of 130 US-based executives and business leaders representing organizations with more than $1 billion in annual revenue found that the percentage of organizations deploying artificial intelligence agents quadrupled from 11% to 42% in just six months.
At the same time, Rahsaan Sears, principal and aIQ program leader at KPMG US, says the third quarter marked a fundamental shift in how people approach technology. As more people actually worked with these tools, the “fear factor” disappeared and was replaced by “cognitive fatigue,” she said. Reflecting this sentiment, a KPMG report highlights that employee resistance has fallen dramatically, from 47% last quarter to just 21% now. More than half of employees now embrace or are actively embracing AI agents. Technology leads the way, with 95% reporting the use of agents to improve productivity, followed by operations management and risk management.
“People on the ground” are reassuring, but tiring.
Sears said executives are saying that as more and more employees engage with the technology, either through enterprise tools or by participating in organizational proofs of concept, “we'll start to see where it's possible.” However, their maturity level has not yet reached a stage where they can fully replace human workers. We need “a human being in the loop, or a human being in the loop.”
A crucial insight for Shears is the changing relationship between humans and AI in the workplace. Sears describes this technology as “infant.” That is, although they are capable of impressive feats, they are still immature and require context, guidance, and supervision. “This isn't infantile in all fields; software development, for example, is much more advanced. But most enterprise applications still require human intervention, she observed.”
Sears believes that the continued need for human skills will make employees more comfortable with AI, and that AI is an enabling tool rather than a replacement for AI. “There's a constant need for human interaction, and I think people find that comforting,” Sears said, highlighting the unique skills needed right now, such as critical thinking, questioning and adaptability. She said she believes “renaissance skills” will become increasingly important, but clarified that that doesn't mean the workforce will be filled with poetry majors. Still, in order to become a stakeholder, it is important to have the skills to think and ask questions.
Asked if people have ever experienced an AI tool being bad or lacking a return on investment, Sears said people have gone from expecting AI to do the job as well as someone with more experience to understanding that while it can work much faster than humans at many things, like a toddler, it can also cause great damage if not closely monitored.
Rethinking success and ROI in the age of AI
Both KPMG and Sears argue that traditional business metrics are insufficient to capture the transformative impact of AI. Research shows that 78% of leaders say traditional KPIs lack much of the value of AI. Sears said the same goes for ROI and the much publicized failure of many AI pilots to achieve ROI. “I don't think the traditional measures that we've been looking for can tell us everything, because we're not going to do very long analyzes and we don't know how to necessarily measure all the right indicators. I'm very interested and intrigued to see how this comes to fruition,” she said, adding that KPMG is looking at a wide range of signals to evaluate the results of AI.
KPMG data reflects this shift. Organizational leaders are now tracking productivity (97%), profitability (94%), and quality improvement (91%) as evidence of AI's business impact, even as broader company outcomes continue to evolve.
Aiming for a new type of workforce
The transformation of the workforce is now firmly underway, and Sears sees hope, especially for entry-level workers who, despite being “digital-first,” face higher expectations for skepticism, critical thinking, and adaptive reasoning.
When asked if this would change the way entry-level work feels or looks, Sears said it's a bit subtle. Because we're seeing a trend among younger workers who grew up with social media and constant access to their iPhones to “trust” their devices and technology. In the case of AI, AI is “in its early stages of maturation, so we need to be more skeptical. This is a different kind of relationship than we've had in the past from a digital interaction perspective.” This is consistent with KPMG's findings that 56% of leaders expect to reshape entry-level hiring within the year. Asked if employers need to take a more critical look at entry-level jobs, with less menial tasks, Sears said it's already happening: “We're seeing it.”
