New research presented to Commonwealth Bank employees is challenging the assumption that faster AI adoption automatically means better outcomes, with academics urging organisations to build deliberate pauses into AI-assisted workflows.
Renée Richardson Gosline, senior lecturer at MIT Sloan School of Management and head of MIT's Human-First AI Lab, says some of the most useful AI systems are those that use friction deliberately, prompting people to test assumptions rather than accept an AI-generated answer outright.
The goal, Gosline said, is to stop workers defaulting to the easiest available answer, particularly where speed could come at the cost of accuracy or judgement. She argued organisations need to avoid both extremes in how staff relate to AI tools.
"We have people who are AI sceptics and may be missing out on some of the value that be created by these tools," Gosline said. "At the same time, we have some folks who talk about AI as if it's a panacea."
That same imbalance shows up in Australia's broker channel. A Connective AI Readiness Report of 300+ brokers found 86% see AI as essential to staying competitive — yet 65% have no clear AI strategy, and over 40% have no governance in place at all.
Back inside CommBank, chief AI scientist Mary-Anne Williams said the most valuable AI use cases were those that freed staff to focus on higher-value work rather than replacing human judgement altogether. AI was suited to fast data processing, she said, but not tasks requiring deeper contextual understanding or empathy.
Gosline pointed to similar limits in AI's capacity for original thinking, saying it tends to reflect "the mean" — the already-established path — rather than genuinely new ideas. She encouraged staff to "practice acts of inconvenience to engage their critical thinking, their curiosity, their exploration."
That same principle extends beyond the bank's own workflows. Cotality's AI in Housing 2026 Report found 75% of homebuyers expect AI in the mortgage transaction — but most still want human-verified decisions.
The session drew more than 1,200 CommBank employees, a turnout CommBank chief behavioural scientist Will Mailer said "reflects a deep curiosity amongst our people about how to best apply AI in their daily work and in their teams without sacrificing human skills."
The research forms part of CommBank's ongoing collaboration with MIT Sloan on responsible, human-centred AI, an initiative the bank formalised through its Human-Centered AI unit in 2024. Chief technology officer Rodrigo Castillo said the findings would help shape a more human-centred approach to the workforce going forward.
Get the hottest and freshest property and mortgage news delivered right into your inbox. Subscribe now to our FREE daily newsletter.
