The real cost of a workplace where nobody wants to go first

Most organisations say they want people who will speak up, challenge assumptions, and find better ways of working, writes Yasmin London.

We write innovation into company values, encourage experimentation, and tell people to bring ideas to the table. But people work out pretty quickly whether we actually mean it.

They see what happens when someone questions a decision their manager has already made, when an experiment fails, or when a colleague points out that the new AI system everyone is excited about might be creating another problem. Over time, people learn what is safe to say, what gets rewarded, and what’s better left alone.

That has always been costly. In the AI era, the stakes are higher.

As AI becomes part of how work is allocated, monitored, assessed, and performed, organisations need people who are prepared to question what they are seeing, flag problems early and use their own judgement. The risk isn’t simply that employees will be too slow to adopt AI. We should also ask what happens if they become too willing to accept it.

AI is changing work, which means it’s changing the rules

Once an AI system changes how somebody works, how their performance is measured, how decisions about them are made, or how much control they have over their work, it becomes an HR issue, too.

In NSW, the Work Health and Safety Amendment (Digital Work Systems) Act 2026 received assent in February, bringing algorithms, AI, automation and online platforms into the state’s approach to digital systems at work. Key provisions are yet to commence, with SafeWork NSW developing supporting guidelines.

Safe Work Australia has also released guidance on managing AI and digital technologies at work, including the potential for these systems to introduce or increase psychosocial hazards. Low job control, excessive workloads, and poor consultation during organisational change are already recognised psychosocial risks. AI gives us new ways for those risks to show up.

If AI allows someone to work faster, does their workload reduce, or does the expectation simply increase? If a system monitors performance more closely, what happens to autonomy? If a role is redesigned around AI, has the person doing that job had meaningful input?

These questions belong in our AI strategies too.

The more capable technology becomes, the more we need people to think

Imagine an employee notices that an AI-supported process is producing questionable results. The system is fast, the dashboard looks impressive, and their manager is enthusiastic about the productivity gains.

Does that employee feel able to say, “Can we check this?”

As we automate decisions, recommendations and workflows, human judgement becomes particularly valuable when technology misses context, produces an error or creates an outcome nobody anticipated. Yet when people are under pressure to move faster and demonstrate AI adoption, challenging the system can feel like slowing everyone down.

AI capability, therefore, cannot stop at teaching people how to use the tools. We need people who can interrogate information, recognise when something doesn’t look right, disagree constructively and make decisions without perfect information.

The World Economic Forum’s Future of Jobs Report 2025 found AI and big data among the fastest-growing skills, but analytical thinking remains the most sought-after core skill. Resilience, leadership, creative thinking, and curiosity and lifelong learning also feature prominently.

There is little value in giving people extraordinary technology if they become less confident using their own judgement.

Who experiences the change?

Not everyone will experience AI transformation in the same way.

The International Labour Organisation reported in 2026 that 29 per cent of employment in female-dominated occupations has some exposure to generative AI, compared with 16 per cent in male-dominated occupations.

For HR leaders, workforce planning and diversity, equity and inclusion (DE&I) strategy cannot sit separately from AI strategy.

Who gets access to reskilling? Who gets invited into emerging roles? Who is involved in redesigning jobs? Are we looking at whose careers could be accelerated by AI as closely as which tasks could be automated?

The people experiencing the greatest change may also see consequences that weren’t visible when the implementation plan was written. If they don’t feel comfortable raising those concerns, organisations lose valuable information when they need it most.

Creating the conditions for first movers

We often tell people to be courageous, innovative, adaptable, and willing to speak up. But individual courage can only take us so far. If questioning decisions is risky, mistakes are punished or challenging the status quo comes at a personal cost, people become more cautious. We cannot simply ask people to become first movers. We need to create the conditions that allow them to be one.

I think about those conditions through three priorities: protect the human, preserve the mind, and power human connection.

Protect the human means considering what is happening to workload, autonomy, role clarity, job security, and psychological safety as work changes. The World Health Organisation includes excessive workloads, low job control, job insecurity, discrimination and exclusion among workplace risks to mental health.

Preserve the mind means developing the judgement we are going to need more of, not less. If AI can give us an answer in seconds, our people need critical thinking, curiosity and discernment to decide whether it is a good one.

Power human connection means creating workplaces where information moves effectively between people. Employees need to be able to challenge a decision, admit they don’t know, collaborate, share an early concern or say, “Something isn’t working here” without wondering whether it will count against them.

When those conditions are present, people are more likely to exercise judgement, raise what they notice and take responsibility for acting on it.

That is how we create more first movers.

The real cost when nobody wants to go first

Being a first mover isn’t about being the loudest person in the room or taking unnecessary risks. It is being willing to notice what others have missed, ask a better question, act without perfect certainty, and take responsibility for what happens next.

Leaders influence whether those behaviours grow or disappear. We cannot ask people to experiment and punish every imperfect attempt. We cannot tell them to speak up if experience tells them agreement is rewarded.

When people stop going first, problems are raised later, assumptions go unchallenged and capable people start waiting for permission rather than exercising judgement.

AI makes that an expensive vulnerability.

For HR leaders, the opportunity is bigger than helping people adopt another generation of technology. We can create workplaces where people use increasingly powerful AI while retaining the capabilities organisations will continue to need from them.

Protect the human. Preserve the mind. Power human connection.

The organisations that get AI right won’t just be the ones with the best technology. They’ll be the ones with people who still have the courage to question it, the judgement to know when something isn’t right, and the agency to act.

Yasmin London is a former international gold medallist swimmer and decorated police officer. She is now a globally recognised keynote speaker, event host, podcast host, and founder and CEO of The First Movers Co.

Want to see more stories from trusted news sources?
Make HR Leader a preferred news source on Google.