An Australian bank has come under fire after dismissing a long-serving employee who unknowingly trained the artificial intelligence system that ultimately replaced her. Kathryn Sullivan, 63, was made redundant in July after 25 years with the Commonwealth Bank of Australia (CBA), following months of work scripting and testing responses for the bank’s new chatbot, Bumblebee AI. “Inadvertently, I was training a chatbot that took my job,” she said, describing her dismissal as “completely shell-shocking.”
Sullivan, who had supported technological improvements in customer service throughout her career, revealed she and colleagues were blindsided by the decision, accusing the bank of treating them like “numbers.” Her concerns were compounded when CBA failed to respond to her queries for more than a week, only later admitting the AI rollout had been mishandled. The bank acknowledged that its decision to cut 45 jobs had “not adequately considered all relevant business considerations” and apologised to affected staff.
While CBA reversed some redundancies and offered roles back to impacted employees after customer calls surged, Sullivan opted for redundancy, saying the new role offered lacked the security of her original position. The bank faced further criticism for simultaneously hiring about 100 new workers in India just weeks after cutting more than 300 jobs in Australia. Despite the controversy, CBA reported a $10.25 billion profit last year and continues to expand its AI operations, recently announcing a partnership with OpenAI to fight scams, fraud, and cybercrime.
The case has sparked wider debate in Australia about the risks of AI adoption in the workplace. Speaking at a national AI symposium, union leaders and policymakers warned that workers must be protected from abrupt technological disruptions. Reserve Bank governor Michele Bullock stressed the need for retraining and education, while Assistant Productivity Minister Andrew Leigh argued that “technology should serve people, not the other way around.” Labor senator Tim Ayres added concerns about Australia relying too heavily on foreign-developed technology rather than shaping its own digital future.
For Sullivan, the experience underscores the need for stronger safeguards. While she acknowledged AI’s potential, she urged for regulations to ensure it complements rather than displaces human workers. Her story has since become emblematic of the challenges Australia faces in balancing innovation, economic growth, and the rights of employees in an increasingly automated world.
