AI in action: Fighting fraud, personalising customer experience and streamlining compliance in banking

AI is no longer just hype or boardroom buzz. It’s undeniably driving measurable transformation across banking in Australia and globally.

But, as more banks adopt AI, many uncover challenges in moving from pilots to real value.  As explored in our previous blog, AI initiatives demand strong foundations in data, strategy, governance and culture to succeed in production and at scale.

With these in place, banks can begin to unlock AI’s potential across the front, middle and back office. But the key is not deploying AI anywhere and everywhere. Banks need to focus their AI efforts on real business problems where it can deliver a clear, measurable impact.

Fighting fraud

Fraud remains one of the biggest threats to banks and their customers, with Australians losing AUD $2.03 billion to scams in 2024, according to Scamwatch. Fraud tactics are becoming increasingly complex, with criminals themselves turning to AI to exploit vulnerabilities.

• Personalised scams powered by deepfakes or voice cloning.
• Use of LLMs to scrape sensitive customer data to fuel highly targeted and hard-to-spot phishing attacks.
• Generation of realistic-looking official documentation to bypass existing checks.
• The ability to probe fraud detection systems at scale via AI to learn how to bypass them.

While AI has increased threat potential, the good news is that it is also becoming one of the most effective defences in the fight against fraud. As the Australian Banking Association (ABA) Chief Executive Simon Birmingham noted, it is both the “new weapon of choice for scammers” and a “powerful tool” to combat fraud.

Banks can now turn to AI to proactively prepare and support customers to prevent them customers becoming victims of fraud, while also accelerating learning and remediation efforts for fraud instances that have occurred in an holistic and integrated strategy.

• Real-time fraud detection: Analyse millions of transactions and user interactions for anomalies.
• Identity verification: Spotting synthetic or stolen identities through biometrics.
• Deepfake and voice fraud detection: Identify inconsistencies in AI-generated voice and video.
• Customer protection: Flagging suspicious communications and transactions.

At Equal Experts, we’ve helped organisations strengthen their fraud infrastructure, including enabling a global retailer to increase fraud detection by 15% and save USD $7 million in chargebacks and false-positive order cancellations. Read the case study: Beating ecommerce fraud at scale

Personalising the customer experience

Customers are demanding more from their banks, with a recent Forrester report finding that nearly half of Australians want personalised banking experiences. With access to clean, reliable and secure customer data, AI helps organisations move away from a one-size-fits-all service to a more tailored banking experience with the customer at the centre.

• Next best action: Based on an individual customer’s situation, behaviour and preferences, identify the most effective and relevant action for engagement.
• Tailored financial insights and advice: Analyse customer history to support everything from savings nudges and budgeting tips to wealth management advice and retirement planning.
• Smarter product recommendations: Matching customers with financial products such as credit cards, home loans or savings opportunities that meet their needs.
• Customer journey optimisation: Analysing interactions across apps, call centres and branches, to aid understanding of customer preferences.

We’ve seen similar personalisation success beyond finance, including helping a global leisure organisation use machine learning techniques to identify customers most likely to buy memberships, demonstrating a potential £1.6m uplift in sales. Read the case study: The power of personalisation and tailoring to the right audience

Smarter credit and risk assessment

In the past, banks have traditionally relied on income, credit history and repayment records when making decisions on credit and risk assessments, which can limit a customer’s access to some products and services. But AI is enabling organisations to move towards more dynamic, data-driven and inclusive approaches using modelling that is based on the best interests of the customer.

• Real-time data analysis: With access to more data sets across transaction history, spending patterns and management of savings, AI can model risk from a wider range of data points, providing more accurate lending decisions, potentially opening up greater opportunities to those without lengthy credit histories.
• Dynamic risk profiling: With AI, risk models can be updated in real time as markets or a customer’s personal financial situation change, enabling banks to assess risk continuously, rather than just at the start and aid customers to make the right credit choices.
• Systemic risk management: AI models can be used to simulate changes in the economy, such as interest rate rises or rising unemployment, to stress test different scenarios and enable better decision-making.

With support from Equal Experts, a major US-based online marketplace saw the benefit of a modern data-driven approach to credit approvals. By working to optimise real-time access to customer data, the marketplace was able to increase approval rates for a co-branded credit card by 50% and drive revenue growth of over $23 million through incentives and rewards. Read the case study: Optimising credit card underwriting models

Streamlining compliance automation

As a highly-regulated industry and with ongoing scrutiny from regulatory bodies including the Australian Prudential Regulation Authority (APRA), the Australian Securities and Investments Commission (ASIC) and the Reserve Bank of Australia (RBA),  compliance is one of the biggest challenges for Australian banks. It can also be one of the biggest cost centres if mistakes are made, with ASIC recently issuing its biggest fine to date – $240m – to ANZ for misconduct.

AI can help reduce the cost and complexity of identifying and mitigating breaches, fulfilling obligations and controls reporting and ensuring overall compliance.

• Automating regulatory reporting: Extract, clean and structure data from multiple systems and generate compliance reports for regulators.
• Streamlining cross-agency reporting for complex regulations: Anti-Money Laundering and Counter-Terrorism Financing (AML/CTF) and similar statutes involve monitoring obligations, controls and reporting that span multiple agencies and departments. Using AI, these can be streamlined into a low to no-touch process.
• Maintaining policy compliance: Preparing the compliance documents and maintaining ongoing requirements automatically, helping to increase the risk posture and health of the bank.

We’ve already demonstrated the potential benefits of AI within regulatory compliance, creating an AI-powered tool for complaint handling within financial institutions. The demo can analyse unstructured data from multiple customer channels, categorise and triage the complaints before providing the complaint handler with a timeline and plan for next steps, all based on Australian Financial Complaints Authority (AFCA) regulatory requirements. Read more about the AI-enabled complaints handling system

Identify the right use cases with an AI discovery workshop

AI has a wide range of use cases across the front, middle and back office, but how do you know which one will provide the greatest return on your investment? We’ve heard from many CTOs and CIOs in the financial services landscape that identifying the most valuable AI initiatives is their main challenge.

The starting point is identifying a real business problem that needs to be solved and uncovering if and how AI can provide a solution.  Our AI discovery workshops help you identify high-impact AI use cases and validate them quickly, turning potential into real action. We’ve already helped senior technology and data leaders within the financial and superannuation sectors strategically plan their AI initiatives and align their teams with real insights and early testing to reduce risk and wasted development effort.

Learn more about our AI discovery workshops and contact us today to book your session.

About the author

Matthew Waugh is Director of Sales and Business Development for Equal Experts APAC. With over 10 years of experience in sales and business development, Matthew is a lead in strategic initiatives that drive growth across the Asia-Pacific region for Equal Experts. Connect with Matthew on LinkedIn.