CBA appoints Professor Mary-Anne Williams, an AI researcher, as Chief AI Scientist.

      The AAAI Fellow based in Sydney is set to join the UNSW Business AI Lab, taking charge of a team of Distinguished AI Scientists as part of Commonwealth Bank’s latest expansion in frontier AI.

      On Monday, Commonwealth Bank of Australia appointed Professor Mary-Anne Williams as its inaugural Chief AI Scientist, selecting one of the nation’s prominent AI researchers from the University of New South Wales following an extensive global search, as the bank described.

      Williams will oversee CommBank's team of Distinguished AI Scientists who specialize in machine learning, responsible AI, AI security, and generative AI. She holds the Michael J Crouch Chair for Innovation at UNSW, is the founder of the UNSW Business AI Lab, and serves as the deputy director of the UNSW AI Institute. Being a Stanford Fellow and a Fellow of the Association for the Advancement of Artificial Intelligence, she also features on Robohub’s Top 25 Women in Robotics list. Notably, her Social Robotics RoboCup team from Sydney secured the world championship in 2019.

      Her published research over the last 18 months has centered on how organizations can effectively manage and coordinate fleets of generative AI agents at scale, a topic for which CBA is now employing her expertise in a banking context.

      While the Chief AI Scientist role is a new addition at CBA, the overall expansion strategy that it integrates with is not. Ranil Boteju took on the role of Chief AI Officer in November 2025 after previously serving as data chief at Lloyds Banking Group. In February, CBA announced a $90 million initiative to develop an AI-ready workforce.

      Earlier this month, the bank launched a Technology Hub in San Francisco to co-locate its engineers with AI laboratories, in addition to an existing facility in Seattle. The bank is also forming partnerships directly with Anthropic, AWS, Microsoft, and OpenAI, rather than going through a systems integrator.

      Externally, the cumulative efforts have had a noticeable impact, with CommBank ranking fourth globally and first in the Asia Pacific region according to the 2025 Evident AI Index, which evaluates banks on their AI maturity. Boteju remarked that Williams possesses a unique blend of academic leadership, commercial savvy, and profound technical knowledge that will propel our AI strategy forward.

      Williams stated that CBA has built one of the most advanced AI capabilities among banks globally, emphasizing her intent to further understand the societal implications of AI while fostering responsible AI innovation.

      This hiring strategy mirrors that of major U.S. technology and financial firms over the past two years, targeting senior AI leaders directly from academia instead of from rival banks or foundational labs. CBA’s announcement indicates that they have intentionally adopted this approach, believing that the critical aspect of the AI stack a large bank cannot outsource is linked to the need for frontier-research expertise combined with internal organizational authority, which is more readily sourced from a university than from a current banking position.

      The context in which an Australian bank is making this decision is significant. Australia’s securities regulator, ASIC, is part of a coordinated initiative with financial regulators to oversee frontier AI risks to the banking system, a focus initiated by the Bank of England, the Federal Reserve, and the U.S. Treasury after the release of Anthropic’s Mythos cybersecurity model demonstrated zero-day discovery capabilities on an industry scale.

      The supervisory issue to address in the coming year is determining how much understanding of frontier research major Australian banks need to maintain internally to effectively manage their own AI risks.

      Based on the last nine months, CBA’s stance is that this understanding must exceed what foundation-model partners alone can provide.

      The broader competitive landscape involves the recalibration of AI within financial services. Earlier this month, Anthropic released ten financial-services agent templates, integrated Moody’s data into the Claude workspace, and established distribution through Microsoft 365 and Snowflake. Companies like SAP, Microsoft, and OpenAI are similarly promoting variations of agentic banking tools on a similar timeline.

      By establishing its own team of Distinguished AI Scientists, the bank positions itself to mitigate risks related to the integration layer that will be essential for connecting external products with its own data and regulatory frameworks.

      Details regarding Williams’s start date and specific reporting structure at CommBank have not yet been made public. Her research activities at UNSW will continue concurrently, following the usual pattern for academic leaders transitioning into industry roles at this level.