iscover 
With the entire mortgage industry working to identify and stamp out loan fraud, lenders are increasingly using centralised repositories of information to identify and share suspicious loan applications, according to credit agency Equifax.
According to media reports, lenders may have lent up to $3 billion in loans to borrowers based on fraudulent applications. While the majority of focus has been on organised criminal organisations (such as those believed to be behind the Penthouse Syndicate), the Equifax Fraud Index Report – 2025 Year in Review report recently revealed a 25.5 per cent increase in first-party fraud (where individuals manipulate their own loan applications).
According to Equifax, this trend may be linked to cost-of-living pressures, combined with the growing availability of AI tools that make it easier to falsify documents.
Equifax also noted a rise in money muling, whereby individuals were transferring illicit funds on behalf of others, aiding in money laundering, and obscuring the original source of the funds (up a whopping 90.9 per cent in 2025, accounting for 14.6 per cent of all fraud listings). All of this has been increasingly supported by the widespread availability of artificial intelligence (AI) tools that make falsifying documents easier than ever.
But how, exactly, are these problem applications being tracked?
A key player in the fight against fraud comes through Australia’s largest Known Fraud Exchange, powered by Equifax, which includes data from 150 financial institutions.
Speaking to Broker Daily, Tehani Legeay, general manager of digital identity and fraud services at Equifax, explained that the platform operates as a collaborative industry defence.
“About 150 of Australia’s lighthouse financial services institutions share known or highly suspicious application fraud data with us as a central place,” Legeay said.
“It will include everything that is on an application form: your identity, what product you’re applying for, and the supporting information that you provide to justify your application.
“Whenever any of those lenders or financial services businesses go to assess a loan, they will search that central repository of industry-known, or highly suspicious data, as part of their overall assessment of every loan.”
When a lender flags a file, they can also pinpoint exactly where the application originated.
“Whenever one of our members lists a known fraud or a suspicious application, they will include a range of data elements, and one of that is channel,” Legeay said.
“We have a range of channels that the lender can specify, and one of them is broker.”
However, Equifax data shows that third-party fraud listings account for less than 3 per cent of all fraud listings, with 31.6 per cent coming from first-party channel in 2025 (up from 26.2 per cent in 2024).
While the exchange is powered by extensive data sharing, Legeay stressed that the system relies on human oversight rather than automated rejections.
“Typically financial services institutions will use an application to search the database,” she said.
“But the industry never declines on a match alone. There is always someone reviewing with a fine-tooth comb every aspect of an application.”
This human review prevents false positives on common names, she said, while ensuring genuine red flags are thoroughly vetted.
“They might see that it has been previously flagged, and then they will confirm that it’s been listed on the exchange,” Legeay told Broker Daily.
The database also relies on individual institutions catching new or first-time offenders before sharing that intelligence with the wider market.
“It might also happen that I call these ‘patient zeros’ – someone who has never defrauded anything before,” Legeay said.
“They try to apply with Bank ABC, but get caught out by the bank teams because of their own internal systems or processes. Bank ABC would then list that on our exchange. It’s a combination of ways that you can get identified as a fraudulent application.”
Smaller loans increasingly attractive to fraudsters
However, while larger-ticket items – such as home loan, commercial loans, and car loans – have been traditionally targeted by fraudsters (there was a 112 per cent surge in fraud for big-ticket credit items like vehicles and commercial assets, according to Equifax’s most recent fraud report), smaller amount loans – such as personal loans, lines of credit, and credit cards – are also increasingly being used by those seeking to commit fraud.
According to Equifax, this may be a result of AI being able to automate the process of fraudulently obtaining smaller-ticket products (which may be approved automatically, for example).
Legeay explained that technology has changed the economics of smaller-scale fraud: “As technology advances and becomes readily accessible, perhaps it makes sense for fraudsters, whoever they are – whether it’s first party or third party – to automate the process of fraudulently obtaining lower-value products. They’re now able to spend less time and effort in obtaining lower-value products.
“So, it’s been super interesting to watch how technology has shifted that landscape, especially in the last two years.”
While talks are believed to be ongoing to create a similar fraud exchange forum to foster industry-wide data sharing and collaborative defence in the broking industry, there is growing investment by aggregators and lenders alike to identify AI-generated documentation. For example, the Commonwealth Bank of Australia (CBA) recently announced the development of an AI system designed to help detect fraud.
According to CBA, the new AI agent identifies emerging threats and determines how it can disrupt them, operating round the clock to continuously monitor activity.
The bank said detection rules are then reviewed and approved by CBA’s fraud analytics team before being implemented.
CBA executive general manager James Roberts said the new technology forms part of a $1 billion commitment to strengthen the bank’s fraud and security capabilities.
“When suspicious patterns are identified, the system quickly assesses their severity, analyses context, and proposes new detection rules to help intercept them. The new agent goes beyond traditional AI by not only rapidly identifying new threats but also determining how it can seek to disrupt them,” he said.
“The agent operates around the clock, continuously monitoring activity and adapting to emerging threats.”
Docuscan – an Australian AI-driven platform – also recently announced the launch of FraudX, a new AI-powered fraud detection solution designed to help brokers and lenders uncover financial fraud hidden within document metadata.
Rather than focusing solely on visible content, the platform interrogates metadata – including file creation details, edit histories, and embedded digital signatures – to identify inconsistencies that may signal tampering.
FraudX is designed to detect not just what appears on a document, but what sits behind it – an area that brokers, lenders, and credit teams have historically struggled to assess.
[Related: Broker fraud may be low, but the stakes are high]
Want to see more stories from trusted news sources?Make Broker Daily a preferred news source on Google.