Building a future-fit organisation requires an equal balance of AI and digital-enabled skills and human-centric capabilities, such as accountability, critical thinking and creativity. Here’s how ANZ Bank is striking that balance.

Editor’s note: Last week, we published an article outlining how HR can help to protect cognitive capacity in the age of AI. This week, we’re following up with a case study example of how ANZ bank is protecting these critical human skills in their own workforce.

As generative AI tools become embedded in everyday work, concerns about the impacts to our critical thinking capabilities are rising. 

According to Arun Pradhan, General Manager of Learning and Development at ANZ, it’s a mistake to conflate the symptoms of cognitive decline with the cause.

“If you think about back in the day, people blamed TV for making us too passive, or the web for shortening our attention spans, or calculators for eroding our math skills. The common thread across all of those things is humans,” he says.

“What we’ve learned through behavioural science is that we tend to try to reduce cognitive load and take shortcuts wherever possible. So now we’re just applying that same humanity to AI by finding ways to put less strain on our minds. I don’t believe AI is causing it – rather it’s humans being human.”

From Pradhan’s perspective, AI is “a small part of a broader issue”. One that includes declining tolerance for boredom and growing disengagement from deep thinking.

“I often talk about how valuable being bored is, because being bored is a gateway for creativity, thought and reflection. It’s very hard to be bored today.”

However, despite AI’s potential to supercharge human capabilities (if used correctly), Pradhan concedes that he does worry about the level of critical thinking taking place in society today. 

“Given the state of our technology, you’d think it would be quite high, and yet it’s amazing to see sections of society retreating. Technology is, in some cases, adding fuel to the fire.”

AI as crutch or thinking partner

Pradhan references a quote attributed to Scottish novelist Andrew Lang: “Most people use statistics like a drunk man uses a lamppost; more for support than illumination.”

“That quote reminds me that, yes, you can use AI as a crutch or a substitute, but you can also use it as a thinking partner.

“One of my early use cases for AI was when ChatGPT first introduced the conversational audio feature. I found that going on walks and chatting to it totally changed my relationship with it. I wasn’t getting it to do work for me – I was trying to nut something out, and I could talk with a whole repository of knowledge to think something through.”

He applies the same thinking to everyday professional practice. Before engaging with AI on a topic that matters, he deliberately forms his own view first.

“Before I ask AI a question – if it’s something I care about – I try to form an opinion or at least set some boundaries around what I’m thinking. Then I’ll ask the AI and have a conversation with it.

“I don’t see that as very different from talking with an intelligent, knowledgeable person who sometimes gets things wrong. You can become complacent if you rely on them too much, but ultimately you still need to maintain your own habits and practices.”

“If you think about back in the day, people blamed TV for making us too passive, or the web for shortening our attention spans, or calculators for eroding our math skills. The common thread across all of those things is humans.”

Building AI and human-centric skills through the ANZ Academy

Critical thinking and judgment skills matter even more in a high-accountability environment, such as a bank.

“With AI, everyone has effectively become a manager, even if they don’t have ‘people’ reports. I encourage using AI in a similar way to how you’d lead a team. 

“When you lead, you’re delegating work, but not accountability. You’ll look at your strengths and gaps, and the strengths and gaps of your team members. Then you’ll delegate, and trust them to get things done but, ultimately, you need to understand and own the work your team produces. 

“Accountability becomes a critical skill. Whatever AI produces – just like whatever my team produces – I’m the one on the line. So I encourage people to be deliberate about the partnership. 

“It’s not just about critical thinking; it’s about prioritisation, commercial acumen and understanding where to invest your attention.”

This philosophy underpins the design of the ANZ Academy, the bank’s enterprise-wide capability-building program. AI capability is a core pillar, but not in the narrow sense of prompt engineering or tool adoption.

Rather than solely focusing only on how to use AI, Pradhan and his team are focused on strengthening the human capabilities that must sit alongside it.

For example, the Academy’s next major campaign, rolling out this year, shifts toward responsible use and risk-aware decision-making.

“The fast follow after that learning is [a module on] critical thinking, and that’s intentional,” says Pradhan.

This sequencing helps employees understand where human judgement must sit alongside AI, and the boundaries between the two, he says.

Another way ANZ is using AI to enable critical skills is through AI-generated simulations that give employees a safe, realistic experience of real-world scenarios, followed by AI-generated reflections on the skills they need to strengthen to be better prepared if those situations arise.

“Some of the customised simulations include dealing with a customer when something goes wrong, and you can choose different personality types. The one that seems to work best is the defensive customer, where someone borders on [displaying] abusive behaviour. 

“There are other scenarios too, like a tech leader trying to explain an AI strategy to the board, or coaching a team member around an ethical dilemma.

“These examples highlight the difference between knowledge and skills. People often conflate the two, but knowledge is largely about memory and information. Skills are about practice – you can’t develop a skill just by being told something. You have to work at it over time.”

A key feature of the learning program is its emphasis on teams rather than individuals.

“The structure is often a five- or six-minute video that says, ‘Here’s a concept, here’s a particular technique.’ From there, it’s typically a reflection on how your team is using something like this.

“Then there’s experimentation together, either in the room or virtually, and then it’s about committing to a habit. Not just saying, ‘I’m going to use this,’ but mapping the moments that matter and identifying when you’ll actually do it.

“That’s where they commit as a team. Social proof helps move the bar, and there’s an opportunity for people to challenge each other.”

Pradhan is also exploring the idea of AI “false flags” – similar to phishing simulations – to deliberately challenge over-reliance.

“Sometimes AI should throw out something that’s complete rubbish – and then show it,” he says. “It’s about forcing people to pay closer attention, in a safe way.”

“I found that going on walks and chatting to it totally changed my relationship with it. I wasn’t getting it to do work for me – I was trying to nut something out, and I could talk with a whole repository of knowledge to think something through.”

Habits that protect thinking

At the heart of this approach is the recognition that critical thinking can’t be sustained constantly, but must be deliberately applied at key moments.

Pradhan suggests asking questions such as:

• How can we identify the moments that matter for our people to think critically, be curious, challenge assumptions and stay engaged? 
• And, since nobody can maintain that level of cognitive focus for long, how can we nudge people to ‘switch on’ for those key moments?

He also suggests building specific habits into your processes to ensure critical thinking is layered in.

“One of my habits is that when I’m really confident about something, that’s the cue. I ask myself, ‘What would I see if I was wrong?’ That forces me to look for counter evidence.”

For leaders, this is particularly important in high-volume approval environments.

“I have to do so many approvals, and it’s actually quite easy to just say yes,” he says. “I have to pause, stop and think, and invest some cognitive load. I’m putting my name on this – I’m becoming accountable for it. You’ve got to do the same thing with AI.”

Why people remain the differentiator

Doing the work to determine how human and AI-enabled capabilities work side by side is foundational to developing your organisation’s competitive advantage, says Pradhan.

“Everyone has the technology. What’s the differentiator? It’s people,” he says. 

As AI infrastructure rapidly increases, organisations that invest in judgement, expertise and disciplined thinking will be the ones that stand apart.

“There’s no substitute, ultimately, for expertise,” he says. “Expertise changes with AI, but people still need to understand how things work, and know when to question and challenge.”

In ANZ’s case, that means treating critical thinking not as an abstract ideal, but as a capability that can be practised, reinforced and embedded into everyday work – one habit, one team and one decision at a time.

Gain clarity and stay ahead of AI-driven change with the knowledge and tools to keep your HR practices compliant and your organisation protected with this course from AHRI.