Faster, higher, stronger - AI lifts output, but also stress

This article was produced in partnership with Indeed

Artificial intelligence is reshaping workplaces at remarkable speed, but the employees most fluent in using it are also the ones feeling the strain.

Across industries, Australia’s youngest and most digitally confident workers are reporting dramatically lower wellbeing than their less tech-savvy (and typically older) colleagues. And while AI is often blamed for lifting expectations or eroding boundaries, HR leaders say the real pressure points lie elsewhere: rising workloads, fragile culture, uneven capability uplift and a growing reliance on a handful of “super users” to carry the weight of organisational transformation.

These tensions sat at the centre of an exclusive industry roundtable hosted mid-November in Sydney by HRD in partnership with Indeed. The global job search and recruitment platform’s new research report, Designing work that works, for tech and people, reveals that AI adaptability and stress are rising together with workers themselves pointing not to algorithms but to workload, culture and lack of support as the true sources of pressure.

Participants from industries spanning childcare to real estate, and medicine to defence discussed how perfectionist professions struggle to adopt AI, why patchy data quality hampers progress, and how distributed expertise, psychological safety and transparent communication are essential to healthy uptake. At the same time, workers’ deepest needs – belonging, meaningful work, social connection and balanced autonomy – remain largely unmet. The takeaway: AI is neither the problem nor the solution on its own. Sustainable transformation hinges on pairing technological investment with deliberate, human-centred redesign of work so that efficiency gains fuel wellbeing rather than erode it.

Faster, higher, stronger

The Olympic motto above, adopted by Pierre de Coubertin, founder of the modern Olympic Games, captures what AI has enabled many employees to achieve in a few short years. But employees who suddenly find themselves performing at Olympic speeds are of course not used to this accelerated state.

The numbers tell an uncomfortable story. Generation Z workers report low wellbeing levels at 82% alongside 72% AI adaptability. Their millennial colleagues show 77% low wellbeing and 67% AI adaptability. By contrast, baby boomers sit at just 54% low wellbeing despite only 34% AI adaptability. It seems the more skilled you are with AI, the more likely you are to be struggling mentally. In contrast, older generations are feeling sanguine about the impact of AI even though they are yet to learn to use it well.

"Our latest report shows a significant paradox in the Australian workforce. While AI is making work faster and smarter, the most digitally fluent workers are also the most stressed," said Lauren Anderson, senior strategic adviser at Indeed. "It's the youngest, most AI-savvy generations who feel the squeeze hardest."

Running at Usain Bolt levels is a shock for ordinary people, even if it is AI assisted. While those yet to put on their new running shoes are operating at their usual pace and feeling fine about it.

The real culprits behind workplace stress

Before pointing fingers at the algorithms, the data offers a surprising twist. Only 9% of workers blame AI for making work harder. Instead, 60% cite increased workload and 54% point to declining company culture or morale.

What gives? AI makes work easier, but there is now more of it – and company culture is failing to recognise this. Anderson explains the implications: "The takeaway is clear. No fancy AI tool can fix work if the human side is broken."

This finding should give pause to leaders rushing to implement the latest technology. AI becomes a convenient scapegoat for deeper organisational problems when the actual issues lie in how work is structured and how people feel about their workplace.

One participant described the split experience within their organisation. "We've got two groups. We've got groups actually using it like experts, and then all the others.” Early adopters are often the same types of people who embraced technological change in the past – they end up becoming the leading geese in the ‘flying V’ formation, informally mentoring others in their group.

“These are the people who are willing to experiment,” said one participant. But being the lead goose leads to more stress and much more informal workload: one person becomes the subject matter expert, fielding questions and troubleshooting problems while trying to maintain their actual job responsibilities. "That one person becomes really the key driver on the AI piece, especially when rolling out a big piece of tech," one participant said. "It is a lot for one person."

It’s a bit like the kid in school who was an ace speller constantly being asked by classmates the correct spelling for any unfamiliar word – they are willing to help, but can also get frustrated that people don’t think to use a dictionary sometimes.

The solution lies in distributing expertise rather than concentrating it. Creating champion groups that can share the burden and spread knowledge across different teams prevents burnout among super users. But this requires investment in training and communication that many organisations have been slow to provide.

“AI can speed things up, but it’s not a cure-all. Don’t turn high performers into the help desk. Keep the focus on making work feel better and run smoother. The message in the room was simple: reduce busywork, protect the human parts, and show people how to use AI so their day actually gets easier,” said Anderson.

The perfectionist's dilemma

Certain industries face unique challenges when adopting AI. High-stakes sectors like engineering, medical care and aviation employ professionals who have built careers on precision and accuracy.

"Our organisation is full of engineers who want to get it perfect, and so they'll dither and work on things forever to make it perfect," one participant said. "But when it's not critical, we need that mentality of actually, it's okay to test and learn and fail."

This tension between perfection and experimentation creates bottlenecks. The challenge lies in helping perfectionist cultures distinguish between high-stakes work that demands rigorous testing and lower-risk applications where experimentation should be encouraged. A cautious mindset can slow progress in situations where speed would be safe.

The stakes extend beyond individual comfort zones. As one insurance industry observer noted, many companies are trying to move into AI with terrible data quality. Since AI will extrapolate errors in source data, the accuracy of information feeding these systems has become far more important than ever before.

What workers actually want

Perhaps the most striking finding in Indeed's research has nothing to do with technology at all. Only 25% of people feel truly happy, fulfilled and low-stress simultaneously. The top two wellbeing drivers for Australian workers are a sense of belonging and feeling energised by daily tasks. Yet just 23% of Australian workers feel their company genuinely cares about their wellbeing.

The disconnect has tangible business consequences. Employees with low wellbeing are twice as likely to be searching for a job.

Anderson introduced the concept of ikigai, a Japanese term meaning "a reason for being". It represents the intersection of what you love, what you're good at, what your business needs and where you feel genuinely valued. "As an organisation, are we responsible for finding everyone's ikigai? No. Finding your ikigai is ultimately a personal journey — it’s not something an organisation can, or should, hand you," Anderson said. "But our research shows that belonging is a core part of wellbeing, and that’s where workplaces can make a real difference."

The shift in workplace culture runs deep and those associated with modern working styles are often contradictory. Employees want to be trusted to work independently while also knowing that managers will be cognisant of any overworking issues. They want to work from home and protect their personal time from being encroached on by work events, but they also want strong social connections with co-workers.

"Everyone wants to work from home as much as possible. But they've also got this desperate need for more and more personal contact," one participant said.

This contradiction isn't easily resolved through policy alone. Large-budget culture initiatives are being scaled back, yet small, frequent moments of recognition continue to have outsized impact. The teams seeing the most success are those building broad AI capability while reinvesting efficiency gains back into people and connection.

Different industries experience these tensions differently. One participant involved in the childcare sector said wellbeing levels were high, despite limited technology adoption. Office-based knowledge workers report higher stress even with better tools at their disposal. Regional, trade-based employees face external pressures like cost of living that overshadow workplace technology concerns.

“You don’t need a big culture rollout to make work feel better. It’s the small stuff — noticing great work, letting people shape how they do their jobs, making space for real connection. If AI gives you back half an hour, spend a little of it on your people. That’s what actually lifts the vibe,” said Anderson.

The path forward

The research suggests that AI itself isn't the problem, nor is it necessarily the solution. What matters is how organisations implement technology alongside human-centred practices.

Continuous improvement programmes that incorporate AI work best when they create psychological safety to experiment. When workers know they can test and learn without dire consequences, adoption accelerates. When every mistake feels like it could crash a plane, progress stalls.

Transparency around different treatment or conditions for employees becomes essential. People accept valid reasons for differences, but they need those reasons explained clearly. Trust operates as a two-way street in these arrangements.

The organisations finding success are those that view AI as a tool to strip out menial tasks and release creativity rather than as a way to squeeze more output from the same headcount. They recognise that curious super users need support and broader skills uplift across teams. They understand that hybrid work requires new approaches to culture building, even if formal training for leading distributed teams remains scarce.

Indeed's research shows 85% of workers feel optimistic that happiness is possible at work. That optimism represents an opportunity. But capturing it requires leaders to address the human side of work with the same rigour they apply to technology implementation.

The most digitally fluent workers shouldn't also be the most worn out. When they are, it signals that something deeper needs attention, something no algorithm can fix on its own.