Comyn calls deal a key step to stay competitive – but are jobs on the line?
The Commonwealth Bank of Australia has inked a multi-year deal with US artificial intelligence firm OpenAI, pledging to accelerate the use of generative AI across its operations in a move that is likely to impact the the mortgage and finance sectors.
The bank said the partnership – announced on Wednesday – would see CBA engineers collaborate with OpenAI’s team to deploy tools aimed at improving fraud detection, enhancing customer personalisation and upskilling staff in the use of advanced AI systems.
Financial terms of the arrangement were not disclosed. CBA announced the deal alongside its annual results, which showed that broker-originated loans flows are falling in comparison to propritary loans at the bank.
CBA chief executive Matt Comyn framed the deal as a key step in keeping Australia competitive in a rapidly shifting technology landscape.
“To be globally competitive, Australia must embrace this new era of rapid technological change,” Comyn said. “Our strategic partnership with OpenAI reflects our commitment to bringing world class capabilities to Australia, and exploring how AI can enhance customer experiences, better protect our customers, and unlock new opportunities for Australian businesses.”
OpenAI chief executive Sam Altman said the collaboration would “put advanced AI into the hands of more Australians, making it more useful and impactful for people and businesses across the country.”
A dual-track approach: innovation and cost-cutting
While CBA insists the rollout of AI will strengthen its service capabilities, the move comes against a backdrop of ongoing criticism over job losses linked to automation and offshore outsourcing.
In July, the bank confirmed the removal of dozens of call-centre positions, citing the introduction of an AI chatbot as the primary reason for the cuts – the first time it had directly tied redundancies to artificial intelligence deployment.
Two weks earlier, the Finance Sector Union (FSU) accused CBA of “replacing skilled Australian workers with AI systems as well as cheaper offshore labour,” and warned that customer service would suffer if human-to-human contact continued to be reduced.
FSU national secretary Julia Angrisano argued that technological upgrades must involve redeployment and retraining rather than job displacement. “Workers want a tech savvy bank, but they expect to be part of the change, not replaced by it,” she said.
CBA has defended its approach, saying AI frees up staff to focus on more complex customer needs while streamlining simple inquiries. A spokesperson noted the bank had hired more than 9,000 people in the past financial year and currently had hundreds of vacancies across retail banking and frontline services.
Implications for mortgage professionals
For mortgage brokers and lenders, the partnership could signal faster credit decision-making, more sophisticated fraud prevention in loan applications and improved client onboarding.
The use of ChatGPT Enterprise – which CBA staff will gain access to under the deal – could also support internal lending teams with faster document analysis, scenario modelling and compliance checks.
Industry observers say the challenge will be balancing efficiency gains with the retention of specialist lending expertise, particularly in complex mortgage scenarios where nuanced human judgment remains essential.
Globally, OpenAI has signed similar agreements with banks including the UK’s NatWest and US investment giant BNY Mellon. CBA has been building AI capability for years, including sending top engineers to Silicon Valley for placements with technology leaders such as Amazon and Microsoft.
For now, the bank’s latest announcement positions it as a leader in AI adoption within Australian financial services – even as the debate over its impact on frontline banking jobs continues.
What CBA’s openAI deal could mean for mortgage lending
Faster loan processing
-
Automated document checks: AI could read and verify income statements, tax returns and bank statements in seconds, reducing manual processing time for credit assessors
-
Quicker pre-approvals: Generative AI could summarise borrower information and flag eligibility within minutes, speeding up conditional approvals for brokers and customers
Enhanced fraud prevention
-
Advanced anomaly detection: AI models trained on historical lending data can flag inconsistencies in application details, such as mismatched income patterns or doctored documents
-
Real-time verification: Cross-checking applicant data with external databases in seconds could make identity fraud and loan-stacking harder to perpetrate
Personalised loan offers
-
Tailored product matching: AI could suggest mortgage products based on a customer’s financial profile, repayment behaviour and future goals
-
Scenario modelling: Brokers could run rapid ‘what if’ interest rate and repayment scenarios during client meetings using AI-generated outputs.
Broker support and internal efficiencies
-
Policy interpretation: AI could instantly parse complex credit policies, reducing the risk of submitting non-compliant applications
-
Market intelligence: AI summarisation tools could provide up-to-the-minute property market trends and borrowing cost projections
Challenges and risks
-
Skill shifts: Lenders and brokers will need training to interpret AI outputs and spot potential false positives in fraud detection
-
Regulatory scrutiny: AI-driven credit decisions will face close examination to ensure compliance with responsible lending obligations and anti-discrimination laws.
Bottom line
If deployed effectively, AI could compress approval timelines from days to hours, strengthen risk controls, and free brokers to focus on client relationship building. But the human expertise in complex mortgage structuring will remain irreplaceable — at least for now.
