ANZ Group Holdings Ltd. Chief Executive Nuno Matos warned of unforeseen risks tied to artificial intelligence and said he could not dismiss the prospect of extensive job cuts as the bank adopts the technology.
Speaking Tuesday at the Australian Financial Review Asia Summit in Sydney, Matos said the pace of risk creation from AI is exceeding expectations. "At this point in time, it’s creating risks at a much higher pace than what their own builders and developers were thinking," he said.
Matos referenced recent public remarks by leading AI executives - Elon Musk of SpaceX, Dario Amodei of Anthropic PBC, and Sam Altman of OpenAI - who have warned about the hazards of rapid model development and have urged a slowdown. Those cautions, Matos noted, have provoked criticism from President Donald Trump.
"They are saying that the risk of this technology is well above what they expected," Matos said, reinforcing concerns from some of the industry’s most prominent figures.
He acknowledged the positive potential often associated with AI, saying: "We all had this dream that AI would come to improve our lives, increase our capacities and capabilities, find cures to many diseases - and that’s true." However, Matos stressed that without sufficient controls the technology could also cause harmful outcomes. "Without 'enough guardrails and limitations,' it will also lead to dangerous outcomes," he said, adding a stark risk projection: "It’ll attack infrastructure."
On the subject of employment, Matos flagged the potential for AI-driven workforce disruption at ANZ, which employs about 40,000 people. He would not rule out substantial reductions. "I don’t know and I think whoever says something that is certain will lie to you - you don’t know," he said. "Nobody knows."
ANZ’s comments contrast with a more optimistic view presented by rival Westpac Banking Corp. on Tuesday. Westpac detailed how specific AI tools could free employee capacity across mortgages and consumer finance, quantifying the anticipated benefits in operational hours and manual task reductions.
According to a presentation by Andrew McMullan, Westpac’s chief data, digital and AI officer, five agents deployed across mortgages and consumer finance are expected to remove 250,000 manual activities and recover about 150,000 hours of banker capacity annually. Westpac estimated roughly 100,000 hours of annual benefit from automating payslip and bank statement checks within home loans, plus about 50,000 hours in consumer finance.
The contrasting messages from ANZ and Westpac illustrate divergent corporate assessments of AI’s near-term effects on operations and staffing.
Market context note: ANZ-1.17% WBC-0.71%