This product has already been purchased.
- The biggest hurdle in scaling AI isn’t the technical deployment, it’s overcoming natural employee scepticism.
- Rather than relying on top-down corporate mandates, successful companies are achieving massive adoption rates by empowering peer champions, offering transparent reskilling and building psychological safety.
- Rushing to deploy AI as a “shiny new toy” is a mistake – leaders must prioritise strategic alignment over speed.
Ask the leaders of Australia’s largest retail, banking and legal firms the secret to scaling artificial intelligence and their answer has surprisingly little to do with technology. It’s entirely about culture.
Across the board, leaders are finding that executing an IT deployment is easy, but getting staff to actually embrace the tools is incredibly difficult. Overcoming the natural scepticism that comes with top-down corporate mandates requires managing the human element of innovation.
Recognising this, Coles completely sidestepped the traditional IT rollout.
Speaking at CommBank’s Accelerate AI summit in May, Coles CEO Leah Weckert shared how the supermarket giant achieved a remarkable 75 per cent active weekly adoption rate of AI tools across its support centre in just 10 weeks.
Rather than relying on a blanket IT rollout, the company cultivated a grassroots movement.
“We made a decision late last calendar year to roll out ChatGPT and Copilot to all of our store support centre team members,” explained Weckert.
After investing in the right technology to ensure staff had access to “good tools”, Coles put an initial cohort of 500 employees through a period of practical training.
“We put them through intensive training,” said Weckert. “We made them champions in their functions and gave them support to find early use cases. When we went live with everybody… we were able to use them as a group of people who were real advocates and champions around the business.”
CommBank CEO Matt Comyn agreed a cultural shift cannot take root if employees view innovation as a threat. Speaking alongside his Coles counterpart, Comyn acknowledged the palpable anxiety across the broader workforce.
To counter this fear, the bank has taken proactive steps to engage its team, launching an internal platform in March designed to map current roles to the skills needed in the future, giving employees a clear, actionable path forward.
“[We covered] things like transparency, being prepared to communicate on the topic on how we’re using the technology,” said Comyn. “For us, and in any regulated industry, trust is really the permission to be able to scale and this technology, of course, attracts an enormous amount of attention.”
In professional services, the entire culture is historically built around the scarcity of human time. Adapting to AI, argued Sam Nickless, CEO at Gilbert + Tobin, requires tearing up this playbook and fundamentally rewriting what an organisation values.
Cultural messaging becomes critical, he said. Namely, if you tell employees that AI is just there to “cut costs”, they’ll reject it out of self-preservation. If you tell them it is there to elevate their potential, they’ll lean in.
“The basic metrics in a law firm, traditionally, have been leverage utilisation and the average rate per hour… But the concept of charging by the hour when the delivery of the service is no longer based on the scarcity of labour… it’s going to unwind that in a big way,” said Nickless in a panel session alongside Comyn and Weckert.
Acknowledging that the entire industry is still at the beginning of this transition, Nickless noted that both firms and clients must come to terms with these new operating models. Beyond altering how services are produced and billed, the widespread embrace of AI will inevitably disrupt how employee performance is judged.
The stakes for navigating this cultural and operational pivot couldn’t be higher. “For professional services firms, if you get this wrong, it’s existential,” he warned.
The danger of the shiny toy
This deliberate, culture-first approach is exactly what many organisations are missing. Paul Bassat, co-founder and partner at venture capital firm Square Peg, warned against rushing to deploy AI purely for the sake of speed, observing that too many companies are treating it like an IT arms race rather than a strategic shift.
“Everyone right now is focusing on the fast bit rather than the right bits,” said Bassat at Accelerate AI. “It’s not just about speed, it’s not just about this shiny new toy… What are we actually going to use this exciting, shiny new tool for?”
Joining Bassat on a panel, Angus Dawson, senior partner at McKinsey, outlined three fundamental questions CEOs must answer to successfully navigate AI.
“One is, do you have enough of a read on what’s going on in the world and where you’re at as a company?
“Two, have you made a decision about how we get the company in the direction you want to go?
“Three, have you done enough to line up the organisation to be able to mobilise and deliver?”
Dawson observed that right now, leaders are pouring all their energy into the third bucket in an attempt to figure out how to get things moving as quickly as possible.
“I’d encourage you to make sure you’ve thought about the first two questions, because that’s what’s needed to go for value,” he said.
The best organisations, noted Dawson, are stepping back and slowing down, replacing the frantic scramble with a more intense, deliberate focus.
Champion-led approach to AI adoption
Coles’ localised, champion-led approach succeeded where traditional change management often fails because it addressed the human element of technological adoption on several fronts:
Social proof over corporate mandate: When an employee sees a teammate save two hours a week using a new AI tool, it shifts the internal narrative. The sentiment evolves from “management is forcing this on us” to “I want what they have”.
Psychological safety: Learning a new, potentially intimidating technology like generative AI makes people feel vulnerable. It’s far less daunting to ask a “stupid” question to a trusted colleague sitting at the next desk than to admit a lack of knowledge to an external IT trainer.
Contextual relevance: A one-size-fits-all training seminar rarely resonates across an entire organisation. An AI champion in the supply chain team will identify completely different use cases than a champion in HR.
Latest news
Already a member?
Login to view this content