By Katharine Wooller, Chief Strategist, Banking & Financial Services at Softcat plc., examines how AI has the power to continually transform the Financial Services landscape.
Financial services have seen successive waves of disruptive technology: mobile and digital banking, cloud-based solutions, open banking, digital assets on a distributed ledger, and most recently artificial intelligence and large language models. The path to adoption is never smooth, nor indeed linear, and it is typical of our industry that the buzz words and hype can be louder than the actual on the ground reality, where finding a genuine business case for deploying a new technology can be complex.
Coupled with emerging regulation, the appetite for utilising relatively unknown and often unproven technology can be, at best, patchy. As a result, when coupled with slow sales cycles, it can seem like we are, as an industry “feeling our way” in the darkness somewhat.
In my day job supporting over 2500 firms across a plethora of flavours of financial services in innovation with their infrastructure, hardware and software needs, I see a wide variety of uptake and appetite for artificial intelligence. There are parts of pockets of the financial services landscape that are well ahead in terms of their acceptance and deployment of AI. However, it does appear a consensus does seem to be emerging on the most appropriate use cases for artificial intelligence.
All innovation in financial services is driven by a very simple requirement: reduce cost and or risk. I have spent over 20 years supporting a variety of firms, ranging from regulators, central banks, asset managers, challenger banks, insurers and building societies through to niche fintechs and am yet to find an exception. AI offers the potential transformative power to improve efficiency, security and customer experience. Looking at the top use cases shows a true range to how this technology can be deployed.
Riding on the back of previous advances in big data, AI driven trading models analyse market data to optimize trading strategies. Unsurprisingly, hedge funds are able to justify the enormous costs of compute and data storage with potential for improved returns.
AI can analyse patterns to detect anomalies and prevent fraud in real time, with unprecedented sophistication, reducing false positives and improving detection accuracy. Fraud is thought to cost over $485bn globally, as per Nasdaq’s global financial crime report[i], a huge “carrot” to invest in cutting edge technologies to find and stamp out illicit actors. Related, anti-money laundering is a huge concern for regulators globally, so any technology that can improve automation in AML processes, and compliance with regulatory requirements, is welcome.
In the same vein, there are substantial advances in cyber security due to the application of AI, which seeks to improve the identification of security threats, for example through behavioural analytics. There is significant political interest in hostile state actors launching cyber-attacks that could potentially compromise systemic stability and of course cause huge potential reputational damage for firms.
Consumer facing businesses are keen to provide genuinely tailored products and services. AI offers the capacity to provided automated personalised advice, for example with tailored budgeting, or investment advice based on risk tolerance via a robo-advisor.
Perhaps a touch more “big brother” is the ability to let AI support credit scoring and risk assessment by using alternative data, for example location data, spending behaviour and social media content.
Sometimes of less popularity with customers, there is significant update of AI using virtual assistants to handle basic customer queries, reducing the need for (expensive!) human agents. Those caught in an endless loop of having a query not dealt with the logic tree may understandably find these interactions frustrating!
The first recorded example of insurance dates back to 1750BC when merchants created an early form of maritime insurance. Since then, the industry has been using data to price risk. AI provides an improved ability to analyse vast data sets, for example from information provided by the IoT, improving underwriting accuracy by assessing risk factors in real-time or in using predictive analytics to anticipate policy holder behaviour.
Whilst perhaps not as glamorous as some of the other use cases, every financial services firm has to deal with a huge amount of admin, legal contracts, and operational process. This is a huge overhead, riddled with inefficiencies and hamstrung by legacy technology. AI offers the potential to extra and process data, reducing manual work. There’s also an interesting argument about the future of work, and whether AI can take the “less desirable” tasks off our daily plate.
Clearly AI offers great promise, despite understandable concern around its safe and ethical deployment, and a desire for greater regulatory clarity. The vendor landscape is complex, and it is likely that a cluttered marketplace will have some clear winners over the next 2 to 5 years. Early adopters should congratulate themselves on embracing what is a very different way of thinking about data, operations and risk through the lens of AI.
It is likely in the medium-term future that will be few firms who are not deploying AI in some way. Whilst I do not subscribe to the idea that the end of humankind is near, courtesy of the robots, there is a sense of “adapt or die”. Anyone working in the industry, therefore, should seek to understand, and indeed embrace, AI. The future of our industry, to my mind, is brighter because of AI.
[i]https://ir.nasdaq.com/news-releases/news-release-details/nasdaq-releases-first-global-financial-crime-report-measuring?utm_source=chatgpt.com
