ADP’s People at Work 2026 report reveals a paradox that every SME owner managing a team needs to understand before their next AI investment.
If you have introduced AI tools into your business and noticed that some of your team seem busier than ever but somehow less confident about what they are achieving, you are not imagining it. New global research from ADP puts a precise figure on a dynamic that many employers are starting to feel without being able to name it. And for small business owners managing small teams through a period of rapid technological change, the findings are worth sitting with carefully.
ADP Research’s People at Work 2026, based on responses from more than 39,000 workers across 36 markets, finds that half of the global workforce now uses AI at least multiple times a week. One in five workers says they use it almost daily. The technology has moved from a novelty to a routine workplace tool faster than most organisations have been able to adapt their management practices to match.
The paradox in the numbers
Here is the finding that should give every employer pause. Daily AI users, the people most embedded in using the technology, are four times more likely to feel less productive than non-users. Not less engaged, not less happy, just less productive in how they perceive their own contribution.
The explanation ADP offers is intuitive once you hear it. As AI takes over routine, checklist-style tasks, the work that remains for humans becomes more complex, more strategic, and harder to measure in simple output terms. Completing ten tasks in a day feels like a good day. Spending a day working through one difficult, ambiguous, long-horizon problem feels less satisfying, even if it is objectively more valuable. The work has not become less important. It has become less easy to tick off a list.
Dr Nela Richardson, Chief Economist at ADP, describes the shift clearly. “AI is not only changing how work gets done, but also how people feel at work,” she says. “Our data shows that frequent AI users report higher engagement and lower stress. But they also feel less productive. Employers that help workers transition to new ways of working with the technology can better foster a workplace where AI feels less like a disruption and more like a teammate.”
For SME owners, this creates a specific management challenge. If your team is using AI, delegating routine work to it, and spending more time on strategic or complex tasks, the old measures of productivity, tasks completed, emails sent, reports produced, may not reflect what is actually being achieved. The gap between real output and perceived output is where disengagement quietly begins.
What the Australian data shows
In Australia, the ADP data reveals a confidence gap that is even more pronounced than the global average. While 44% of Australian workers use AI at least multiple times a week and 16% use it almost daily, only 13% expect AI to positively impact their job responsibilities in the coming year. That is a striking disconnect between adoption and belief. Workers are using the tools. They are just not convinced it is helping them.
The breakdown by worker type is telling. Knowledge workers, those in roles that are primarily analytical or creative, are the most confident about AI at 22% expecting a positive impact. Skilled task workers sit at 9% and repetitive task workers at just 6%. The workers whose roles are most likely to be changed by AI are the least confident it will be a positive change. That is not a technology problem. It is a communication and culture problem.
Age and gender patterns also show up in the data. Workers aged 27 to 39 are the most frequent AI users, with 28% using it almost daily. Men are using AI more frequently than women on average, with 18% of men reporting daily use compared to 13% of women. For SME owners thinking about how to support different parts of their team through the AI transition, these variations matter.
The upside of frequent AI use
The data is not all cautionary. Frequent AI users report meaningfully better workplace experiences across several dimensions. Daily users report experiencing negative stress at 11%, roughly half the rate of non-users at 23%. They are more likely to say they are part of a work team, more likely to describe being on the best team at work, and more likely to feel their jobs are secure. The link between regular AI use and job confidence is one of the more interesting findings in the report, suggesting that workers who are actively using AI feel less threatened by it than those who are not.
The picture that emerges is of two distinct groups within the same workforce. Workers who have integrated AI into their daily practice and found ways to use it effectively are reporting higher engagement, lower stress, and stronger team connections. Workers who are using it occasionally or not at all are experiencing higher stress and lower confidence. The divide is not between AI users and sceptics. It is between workers who have been supported to use AI well and those who have not.
What SME owners should do differently
Kylie Baullo, General Manager for Australia, New Zealand and Japan at ADP, identifies three things employers need to focus on to close the confidence gap. “The priority now is to close this gap by helping people build confidence in AI, through practical training, clearer use cases and redefining how productivity is measured,” she says. “When organisations get this right, AI becomes not just a tool, but a catalyst for more meaningful and engaging work.”
For small business owners, those three priorities translate into specific actions. Practical training means not just telling people a tool exists but showing them concretely how to use it for their specific role and their specific tasks. Clearer use cases means being explicit about what the AI is there to help with and what it is not, removing the ambiguity that makes people uncertain whether they are using it correctly. And redefining how productivity is measured means moving away from counting outputs and toward recognising the quality and impact of the work being done.
That third one is the hardest, particularly in a small business where performance has often been measured informally through visible activity rather than structured frameworks. But it is also where the most leverage is. A team member who is spending less time on routine tasks and more time on complex work that actually grows the business is performing better, even if their task list looks shorter at the end of the day. Helping them see that, and building the feedback loops that make it visible, is the work of management in an AI-assisted workplace.
The global adoption curve is moving faster than most employers have adjusted for. In Australia, where confidence in AI’s benefit remains low despite high adoption, the gap between using the technology and believing in it is the one that most needs closing. That is not something a new tool will fix. It is something a better-supported team will.
Keep up to date with our stories on LinkedIn, Twitter, Facebook and Instagram.
