Technology can streamline workflows, but it also acts as a magnifying glass for existing dysfunctions. Here are five ways to bridge the capability gap and ensure your team is AI-resilient.

Most organisations are approaching AI as a technology problem. They’re investing in tools, building prompting guides and running AI literacy sessions. All of that is useful, but the organisations that will actually pull ahead aren’t the ones with the best tools. They’re the ones with the best teams.

Here’s what I’m seeing in my work with leadership teams across Australia: the teams that are thriving with AI aren’t the most technically sophisticated. They are the most cohesive. They communicate clearly, trust each other, hold each other accountable and have a shared sense of purpose. In other words, the conditions that made teams effective ten years ago are the same conditions that make teams effective with AI now. Only the stakes are higher.

This is why HR’s role in AI implementation is more significant than many organisations currently recognise.

The AI distraction

There’s enormous pressure on HR right now to respond to AI as an organisational imperative: to have a plan and a strategy. From workforce planning, reskilling, governance, productivity frameworks, the to-do list is long. And while these are all important, there’s a risk that organisations focus so heavily on individual capability uplift that they miss the deeper question: is the team itself ready?

AI doesn’t operate in a vacuum. It operates inside teams. A team with poor communication, unclear accountability and low psychological safety will use AI to produce mediocre work faster. The dysfunction doesn’t disappear; it accelerates. 

According to Gallup’s 2026 State of the Global Workplace report, only 20 per cent of employees are engaged at work globally. In Australia and New Zealand, the picture isn’t much more encouraging at 21 per cent. Disengaged teams don’t suddenly become high-performing because they have access to better tools. What they need first is the foundation that makes any tool, AI or otherwise, work.

One thing I’ve noticed since AI tools became mainstream is that they’ve made team dysfunction more visible, not less. 

“There’s a risk that organisations focus so heavily on individual capability uplift that they miss the deeper question: is the team itself ready?”

AI demands clarity. You cannot be vague with a language model. You need to articulate the outcome, the constraints, the context and the standard. When teams struggle to prompt AI effectively, it’s often not a technology literacy problem. It’s a clarity problem. They don’t have a sharp enough shared understanding of what the work is for, who owns what or what success looks like. These are leadership and culture challenges, not software challenges.

AI expert Justin Williams put it well when he joined me on the Thriving Leaders Podcast: “LLMs are sensational for being able to augment what people do and make things happen more quickly. But it’s the last 15 per cent, that critical thinking, that really requires human intelligence and the diligence to check it for accuracy, for context and for utility. The 15 per cent that AI cannot do is precisely what thriving teams do best.”

Leaders can only hold people accountable when they set clear expectations. When  they learn to be clearer with AI, it helps them be clearer with their people. 

The human skills AI cannot replace

When teams rely heavily on AI for speed and synthesis, the human work that remains is more important, not less. And that work is relationship-dependent.

Thriving teams are built on six interconnected elements: purpose, relationships, accountability, connection, challenge and support and alignment. Every one of those is a distinctly human capability. None can be outsourced to a language model.

The biggest risk I see for teams operating in an AI-rich environment isn’t that people will be replaced by technology. It’s that people will stop collaborating with each other. When answers are available at your fingertips, the impulse to ask a colleague a question, to admit you don’t know something, to have a real conversation about a problem, weakens. And with it, so does the relational fabric that makes teams genuinely high-performing.

It takes vulnerability to say, “I don’t know, can you help me?” That willingness is the bedrock of psychological safety, and psychological safety is one of the most reliable predictors of team performance we have. 

Amy Edmondson’s decades of research on this are consistent: teams that feel safe to speak up, challenge ideas and admit mistakes outperform those that don’t. AI doesn’t create that safety. Leaders and HR do.

What thriving teams do differently in an AI world

In my work with teams, the ones navigating AI most effectively share some common characteristics.

They have a clear team purpose. They know why they exist as a team and how their work connects to broader organisational outcomes. This matters enormously with AI because it determines how they use it. Teams with no clear shared purpose tend to use AI for individual productivity. Teams with a strong shared purpose use it to serve collective outcomes.

They have strong accountability practices. They understand individual accountability, who owns what and to what standard, and shared accountability, where peers hold each other to commitments without needing the leader to chase.

When AI accelerates output, teams with clear accountability structures can review, critique and apply human judgment to what the technology produces. Teams without it tend to accept AI outputs at face value.

“The teams that are thriving with AI aren’t the most technically sophisticated. They are the most cohesive.”

High-performing teams invest in connection. Regular operating rhythms, purposeful meetings and genuine human interaction outside the work are not ‘soft’ extras in a high-AI environment. They’re critical infrastructure. 

Without intentional connection, teams lose the trust and relationship depth that makes honest conversation and collaborative problem-solving possible. They debate well. 

Healthy debate, the ability to challenge ideas, disagree respectfully and think critically together is one of the most underrated team capabilities. 

In an AI-rich world, where information is abundant and outputs are instant, the quality of the human thinking applied to that information is the differentiator. Teams that have built the muscle for productive disagreement will outpace those that default to consensus.

Five things HR can do right now

To enable teams to utilise AI as a capability enhancer, these are the five practices I suggest adopting:      

1. Audit the health of your teams before you invest further in AI tools

Before your next AI capability initiative, ask: do our teams have the relational and operational foundations to use these tools well? A team that struggles with accountability, communication and trust won’t benefit from better technology. Invest in the team first.

2. Design AI capability building at the team level, not just the individual level

Most AI upskilling programs are individually focused. The more powerful intervention is team-based: how does this team use AI together, establish shared norms around AI use and apply human judgment to AI outputs collectively? Build this into your L&D approach.

3. Strengthen psychological safety alongside accountability

These are not separate workstreams. As AI takes over more routine tasks, the quality of human decision-making, problem-solving and creative thinking becomes the competitive advantage. 

Both require environments where people feel safe to think out loud, challenge the status quo and take considered risks. McKinsey’s research found that only 26 per cent of leaders consistently create environments with both high psychological safety and clear accountability. That’s a significant gap, and HR is uniquely positioned to close it.

4. Build clarity as a leadership capability

AI has made clarity non-negotiable. Leaders who are vague about expectations, outcomes and priorities were already creating performance friction. Now that friction is amplified. Build clarity, role clarity, goal clarity, expectation clarity, into your leadership development programs. It is the foundational skill for leading in an AI world.

5. Protect human connection intentionally

As AI handles more of the routine work, the time freed up should not default to more individual screen time. HR can play a role in encouraging teams to reinvest that time in the conversations, relationships and collaboration that AI cannot replicate. Design this into your culture, not just your policies.

The competitive advantage hiding in plain sight

The technology gap between organisations will close quickly. When that happens, the differentiator won’t be which organisation has the best AI. It will be which organisation has the best teams.

The research is consistent on what high-performing teams look like: clear purpose, strong relationships, genuine accountability, meaningful connection, productive challenge and alignment across functions and stakeholders. These capabilities only become more relevant as AI scales.

HR is already in the business of building these conditions. The invitation right now is to be explicit about why it matters more than ever, to connect people strategy directly to AI strategy and to advocate for team health as a business-critical investment, not an HR initiative.

Because when your teams are thriving, your AI investment will too.

Claire Gray is a leadership and team facilitator, executive coach and author of Thriving Teams: When Teams Unite, Align & Achieve. She works with HR leaders and executive teams to build accountability, psychological safety and high-performing cultures. Claire is also the host of the Thriving Leaders Podcast, where she explores the real-world challenges of leadership and team performance.