In July 2025, Commonwealth Bank made headlines when it replaced 45 call-centre staff with an AI voice bot. By August, the bank admitted it was a mistake. The bot couldn’t handle the load, customers faced long delays, and the bank apologized as it moved to rehire or redeploy staff.
Some headlines called it a backflip. Others framed it as an experiment gone wrong. The truth is simpler: this wasn’t a calculated risk, it was a flawed process. Cutting jobs before testing adoption readiness put customers, staff, and reputation at risk.
The real lesson? Smarter AI adoption starts with people, not headcount reduction.
TL;DR
- Commonwealth Bank admitted it was a mistake to replace 45 staff with an AI voice bot in July 2025, and is now rehiring them.
- The failure wasn’t the technology, it was skipping structured adoption steps like workforce readiness and user acceptance.
- The bank is now pivoting to partnerships, research, and training, showing that smarter AI adoption is about augmentation, not replacement.
- Twenty44’s AI/44 Assessment and FOCUSED Opportunity Mapping tools can help organizations avoid the same costly mistakes.
What really went wrong at Commonwealth Bank?
The bank tried to automate routine customer calls with a chatbot, expecting staff to focus on higher-value queries. Instead, call volumes rose, wait times blew out, and existing staff and tem leaders were asked to work overtime to cover the calls. Within weeks, Commonwealth Bank conceded it had underestimated the risks, apologized, and began rehiring affected workers.
This wasn’t an AI failure so much as a process failure. The rollout may have overlooked three essentials:
- Team readiness: Employees hadn’t been properly prepared to work alongside the bot.
- User acceptance: Customers weren’t willing to tolerate poor experiences in exchange for efficiency.
- Workflow nuance: AI was dropped in without understanding where human judgement was still critical.
Or, as Nisha Padmanabhan, Chief Risk officer at the University of Sydney posted on Linkedin: “CBA’s AI misstep isn’t proof that AI doesn’t work. It’s proof that AI without people doesn’t work.”
How is the bank rethinking its AI strategy?
Instead of retreating from AI, Commonwealth Bank is recalibrating:
- Partnering with OpenAI: To co-develop generative AI tools for fraud detection, customer personalization, and enterprise productivity.
- Working with MIT Sloan: To study human–AI collaboration and responsible use:
- Upskilling 52,000 employees: Commonwealth Bank is rolling out ChatGPT Enterprise access while investing in comprehensive training and upskilling programs to build AI capability and promote responsible usage across its workforce.
This shift reflects a wider truth: AI adoption isn’t a tech play. It’s a talent strategy. Building a future-ready human–AI workforce is about augmentation, not replacement.
What’s the bigger lesson for AI adoption in Australia?
Australia is often criticised for being too cautious on AI. Commonwealth Bank’s case shows the real risk isn’t caution, it’s careless adoption.
This is where structured frameworks matter:
- AI/44 Assessment: Would have flagged low workforce readiness, uncovered employee hesitations, and revealed where AI could add value without harming service.
- FOCUSED Opportunity Scoring: Would have shown poor user acceptance and high implementation risk, steering the bank away from frontline automation and toward augmentation opportunities.
“Efficiency isn’t just headcount reduction. True gains come when the strengths of AI and people are combined.”
Closing thought
Commonwealth Bank’s admission isn’t proof that AI is doomed, it’s proof that adoption without structure, readiness, and guardrails is destined to stumble. The bank is now pivoting toward partnerships, research, and workforce enablement. That’s the right reset, but it came at a cost.
The real prize isn’t short-term savings. It’s sustainable advantage, earned by embedding AI responsibly, with people at the centre. Because in AI adoption, the smartest move isn’t speed. It’s starting with the people.
Randy Matheson
Randy Matheson is an innovation strategist with a 25+ year proven track record of turning ideas into digital products. He specializes in working with Generative AI for content creation and using cutting-edge AI tools to create and interact with virtual audiences. He operates out Hamilton, Ontario where he resides with his partner and two large dogs.