How Generative AI can help to manage and reduce Financial Services Scams
On May 23rd, 2024, the Australian Banking Association, in partnership with IBM, will hold an online interactive discussion with ABA members on how AI and GenAI can help manage and reduce Financial Services Scams. In preparation for this session at which I will be presenting, I thought I'd write a brief blog article on the topic. My intention in the following blog article is to demonstrate that the banking industry process for Scam detection and resolution can be significantly improved through the increased use of AI, GenAI, and automation.
Whilst I'm sure my friends, associates, and followers in the Financial Services and Banking Sectors would be most interested in this article, I also think this topic is of interest to everyday Australians. In 2022, Australians lost over $3.1 billion to Financial Services Scams, an 80% increase over the previous year (see here). This would indicate that a large percentage of the Australian population has, at some point, either been impacted by Financial Services Scams or knows someone who has.
So, if you are one of these people or work in the Financial Services or Banking Sector, please read on, as this article will be of interest to you.
Background
On November 2023, the Australian Banking Association (ABA) and the Customer Owned Banking Association jointly announced the Scam-Safe Accord, in which Australian Banks announced new investments to help reduce Fraud and Scams in the Australian financial system. Over 15.4 Billion transactions worth $2.5 Trillion occur within the sector each year. Online e-commerce, which uses digital payment methods, has grown 2X to 5X faster after COVID-19 and is quickly increasing its dominance as the primary sales channel for most BtoC organisations. Increasing e-commerce use with its associated increased use of online payment methods means increasing opportunity for fraud and scams to occur.
The Australian banking sector has committed to improving technology and controls to reduce scams. However, this increase in technology and controls will inevitably lead to more warnings and delays for consumers trying to process transactions online.
Generative AI (GenAI) is a relatively new technology recently popularised by ChatGPT. It provides an opportunity to improve the entire Scam resolution process by not only helping to detect the scams themselves but also by significantly reducing the time it takes to respond to and resolve a scam transaction for the end consumer. This will directly reduce the impact of Financial Scams and improve banking customer satisfaction.
This blog article discusses the current state of this issue in the industry and explains where GenAI can help improve banking customers' scam resolution journey. Hopefully, those in the financial services and payment industries will find this article helpful.
Recent Trends in Digital Banking and Financial Scams and the Australian Banking Industry Response with the Scam Safe Accord
It is clear that COVID-19 accelerated a previously developing trend, which was an increase in e-commerce transactions and, therefore, the volume of digital payments. This increase in the use of digital payment systems has also generated increasing amounts of Scams as Australia's reliance on online commerce increases.
In 2022, Australians lost over $3.1 Billion in Financial Services Scams, whilst globally, the losses were estimated at over $20 Billion USD.
By far, the most common type of scam is Phishing, whereas the most common communication channels used by scammers are text, phone and email (See here), although more money is lost from investment and dating scams than other categories. Phishing scams are where the consumer is tricked into providing sensitive information or login credentials by the communication appearing to come from a trusted source like a bank or government department. Spear phishing is a form of phishing where the text, phone call or email is much more highly targeted using previously gathered personalised information about the targeted individual and may appear to be coming from an individual person trusted by that individual.
In response to this increasing volume of Scams, the banking sector has recently volunteered to support the Scam-Safe Accord. With this accord, Australian Banks have committed to...
While this is a great initiative for the banking industry, it is still relatively new. Banking customers clearly have a number of pain points with the current banking processes for scam resolution, which I will discuss in the next section.
Key Customer Pain Points with Current Bank Scams Processes
Banking and financial institution customers constantly identify the same concerns related to the current scam resolution processes within banks. These include transactions incorrectly classified as scams (false positives), the use of slow and hard-to-use IVR systems to report a scam when the customer is emotionally charged, long wait times for customers to get a status of their case, customers having to constantly chase for feedback and/or having to wait for a long time to get a confirmed outcome.
It is clear that any opportunity to improve the detection of scam transactions by reducing false positives and/or shortening the time to resolution when a transaction has been confirmed as a Scam will significantly increase banking customer satisfaction.
Opportunities for Increased Process Excellence in Bank Scam Operations
A recent analysis by the IBM Institute for Business Value has identified three key opportunities for the improvement of banking scam resolution processes through the use of AI, GenAI, and general automation. Specifically, these include...
The following chart talks through a number of the recommended solutions...
There is clearly a large opportunity to reduce the number of successful scams and to improve customer perception of the scam resolution process currently being driven by banks.
What is GenAI, and as a new technology, why is it useful in the battle against Scams
Traditionally, Artificial Intelligence (AI) has required organisations to build a specific AI model for each individual use case, whether it was machine vision, natural language processing or other. These models, trained on specific use case data, needed to be tuned and constantly monitored for performance, updating the models regularly where performance degrades. Organisations needed to pay for the people and tools necessary to build and manage these individual AI models. As a result, AI was only really economical for higher volume use cases where the costs of AI model development and management were offset by the benefits of using AI.
In 2017, the University of Toronto, in partnership with Google, wrote a research paper called "Attention is all you need". See here if you'd like to read this paper. Without going into detail, this research paper fundamentally changed AI by giving organisations a computationally efficient way of building extremely large AI models. These extremely large models, trained on a variety of data, could be used for various use cases, not just one. As a result, the concept of the "Foundation Model" was born. The Stanford AI team first coined the term "Foundation Model" (see here) to refer to AI models that could be applied across various use cases. These Foundation Models allow organisations to adopt a build once-use many times approach and develop learning solutions. This radically changes the economics of AI by making even lower volume use cases economical. It also allows organisations to use models built by other organisations, reducing even further the investments necessary and improving the economics of AI even further.
Generative AI (GenAI) is any time you use AI to generate content, whether it is text, images, or voice. Large Language Models (LLMs) are a form of Foundational, Generative AI used specifically for text generation. I recently published a video on what generative AI is, which you can watch here if you'd like further information and explanation.
ChatGPT, which was released for public consumption a little over a year ago by OpenAI (see here) and which most people have played with, is a Large Language Model, which is a form of Foundational, Generative AI. ChatGPT demonstrated the capability of these Foundational GenAI models, and ever since, organisations have been racing to adopt this new technology because of its benefits.
The adoption rate of these Foundational, Generative AI solutions has been so fast that the use of ChatGPT has surpassed the adoption of Facebook, the Mobile Phone and even the Internet (see here). I have many clients experimenting with this technology.
Gartner, as well as other organisations, has recently identified Fraud and Scams Detection as one of the top 10 key use cases for GenAI in the Financial Services and Banking sector (see here). Generative AI is potentially beneficial to address Scams in a variety of ways, from providing better detection to improved self-service online channels avoiding IVR wait times, to automated notification of status, to automated routing of cases to reduce resolution wait times, to increasing the number of cases which can automatically be resolved without any human intervention.
The Scams Customer Journey and where GenAI can be used
The diagram below shows the customer journey from identifying Scams to managing the interaction between the bank and the impacted individual to identifying the required actions and speeding the resolution. Along that process flow, more than 10 different opportunities are available to use Generative AI to improve the scams process, reduce cycle times and increase customer NPS. These include but are not limited to...
Conclusions
It is clear that the management of Scams is a major issue for the Australian Banking industry. In response to public demand and the increasing volume of scams, the banking industry has come up with the Scam Safe Accord in an effort to reduce the overall impact of financial scams on individual Australians. That accord mandates that banks should implement a number of checks and balances to reduce the number of Scams that occur and to more effectively address the Scams that are being reported. This is a great start on the part of the banks.
However, as a relatively new initiative on the part of the banks, it is clear that there are significant opportunities for increased process efficiency in this space and for improving customer satisfaction with the Banking Scams Resolution Process.
This article has attempted to demonstrate that AI, specifically generative AI, can provide several ways to increase scam identification and resolution process efficiency and effectiveness and thereby increase bank customer satisfaction.
Dr David Goad is the CTO and Head of Advisory for IBM Consulting Australia and New Zealand. He is also a Microsoft Regional Director. David is frequently asked to speak at conferences on the topics of Generative AI, AI, IoT, Cloud and Robotic Process Automation. He teaches courses in Digital Strategy and Digital Transformation at a number of universities. David can be reached at david.goad@ibm.com if you have questions about this article.
GEN AI Evangelist | #TechSherpa | #LiftOthersUp
7moInsightful take on battling scams through advanced tech solutions. Dr Dave G.
Information Technology Manager | I help Client's Solve Their Problems & Save $$$$ by Providing Solutions Through Technology & Automation.
7moHey mate, that sounds like a fascinating topic! AI and GenAI could really revolutionize scam detection in the financial industry. Can't wait to hear more about it at the online discussion! #excitingtech Dr Dave G.
I grow businesses with SEO-driven content | Helped companies increase organic traffic 2-3x | I share content marketing frameworks that work
7moThat sounds like a super interesting topic. AI and GenAI sure have the potential to make a big impact in scam prevention. Can't wait to read your blog article Dr Dave G.
AI Experts - Join our Network of AI Speakers, Consultants and AI Solution Providers. Message me for info.
7moLooking forward to your insightful article on enhancing scam detection in the banking industry through AI and GenAI.