Most borrowers remain reluctant to use artificial intelligence for home loans, instead preferring the expertise of a broker. But brokers are increasingly adopting the tech, new data shows.
New research from Agile Market Intelligence’s Consumer Pulse survey showed that while artificial intelligence (AI) is becoming more mainstream, only a small proportion of consumers are comfortable using the tech for their home loan journeys.
Instead, Australians still showed a strong preference for human interaction when making one of the biggest financial decisions of their lives.
Conducted between 28 July and 25 August 2025, the Consumer Pulse survey asked 1,500 Australian consumers which avenues they would be most comfortable with when researching the best loan options.
Almost one-third (32 per cent) of respondents to the Consumer Pulse survey said they would most trust a mortgage broker, while 28 per cent would turn to a banker, and 21 per cent would prefer to do it themselves. A further 13 per cent were unsure.
Just 6 per cent said they feel comfortable relying on an AI assistant to research mortgage options, even if it could compare, apply for, and manage their loan.
Why would a consumer use AI?
Among the small cohort who said they would use AI to assist in loan research, speed and convenience were the strongest motivators. Sixty-two per cent cited efficiency as the reason they would turn to AI, while 47 per cent believed the technology could access a wider range of loan products.
A further 41 per cent said they trusted AI to provide more objective results, free from human bias, and 43 per cent admitted they would use AI to avoid perceived sales pressure from brokers or bankers. Privacy was also a concern for some, with 27 per cent of consumers saying they valued the anonymity AI could provide.
The results reflected the dominance of the third-party channel (which currently writes three-quarters of all mortgages in Australia) and the value that mortgage brokers provide consumers.
Michael Johnson, director of Agile Market Intelligence, said the results underline the importance of trust and personalisation in the mortgage process.
“The mortgage broker preference shows consumers value independent expertise and choice when making one of their biggest financial decisions,” he explained.
“The resistance to AI in loan research isn’t about technology capability, but about trust in high-stakes decisions. When it comes to mortgages, consumers want human accountability, the ability to ask questions, negotiate, and receive personalised advice that considers their unique situation.”
Johnson noted that even consumers who are open to AI see it as a complement, rather than a replacement, to human advice.
“Those who embrace AI for loan research are looking for efficiency and objectivity over relationship-building. They want to cut through sales tactics and get comprehensive comparisons quickly – but this doesn’t negate the need for expert guidance in final decision-making,” he said.
Brokers curious about AI, but want more training
While consumers continue to prefer human support, many brokers are increasingly experimenting with AI to free up their time so they can enhance, rather than replace, their client relationships.
Agile’s most recent Broker Pulse: Residential Lending survey (covering 285 broker respondents between 1 and 18 August) found that nearly seven in 10 brokers (68 per cent) already have some level of AI skills.
Almost half (49 per cent) reported knowing the basics, 18 per cent said they were confident users, and 1 per cent said they had fully integrated AI into their workflow.
One in four (25 per cent) brokers told the Broker Pulse survey that they use AI tools regularly and report tangible benefits – from drafting communications and automating administrative tasks to analysing borrower scenarios more quickly.
However, more than a fifth (42 per cent) of brokers said they were curious about AI, but don’t know where to start.
Just 5 per cent of brokers said AI was irrelevant to their business, while 13 per cent have tried the tools, but remain unconvinced of their utility.
Johnson said these findings revealed that the profession is in a “transitional stage” with AI.
“We’re seeing a classic technology adoption pattern. Early adopters are already seeing benefits, while the early majority is forming. The 49 per cent who know the basics need clear pathways to move from awareness to proficiency,” he said.
“Curiosity is a strong signal, but it must be converted into confidence. Vendors need to make it easy for brokers to test and apply AI in ways that feel directly valuable to their day-to-day work.”
Industry ramps up AI rollout
Given the acceleration in the adoption of AI to increase productivity, aggregation groups and large brokerages are rolling out strategies to help their members harness AI responsibly.
Major brokerage Mortgage Choice unveiled its AI strategy last week, including the launch of an AI Academy to provide training and resources for brokers. This follows similar announcements from brokerage brands Loan Market and Aussie, which have also committed to providing AI education and tools for their networks.
The moves highlight a shift in the industry towards equipping brokers not only with digital tools, but with the confidence and knowledge to use them in ways that strengthen, rather than weaken, the human relationships at the core of mortgage broking.
For brokers keen to accelerate their skills, the upcoming AI Edge for Mortgage Brokers program – powered by Captivate Q – will debut in Sydney in October.
The full-day, hands-on workshop is strictly limited to 40 participants and designed for brokerage directors, founders, and executives. It will provide practical guidance on how to integrate AI into daily workflows to save time, scale smarter, and build efficient, future-ready businesses.
