Artificial Intelligence in financial services will become more effective as AI continues to merge human insight with automated analysis, but what is your experience of AI now? In my view financial services AI in the future will affect the following areas:
Transactions: By 2020 at least five percent of all economic transactions will be handled by autonomous software. AI will process payment functions and learn from customer behaviours, through Intelligent Payment Management (IPM). According to Gartner analysts, software will enable a “programmable economy often developed in an open-source fashion and set free on the blockchain; capable of banking, insurance, markets, exchanges, crowdfunding - virtually all other types of financial instruments.”
Savings: AI will help consumers make daily financial decisions and monitor spending. New Personal Financial Management apps use ‘contextual awareness,’ which measures spending habits and online footprints to create personalised advice. Combining pooled financial data with end-user control to offer tailor-made services is a classic AI solution. Similarly, intelligent spending enabled through income prediction based on past behaviour can help users better manage their money. AI has a significant role in developing this kind of analysis.
Making sense of financial transactions for cross-selling opportunities: Machine learning enhances the prediction and classification of information, making it easier for businesses to make data-driven decisions. Financial transaction categorisation services, such as the Categorisation-as-a Service API, CaaS helps financial institutions and Fintechs understand customers’ transactions. With AI, the financial transaction categorisation API automatically categorises user's financial transactions. It transforms and processes all of a users' transactions from their bank, credit card provider, brokerage firm, pension, utility provider or loyalty programme and augments that raw data with categories, merchant data, payment methods and geo tagging. Financial transactions can be understood by all businesses connected to the API and provide language agnostic auto-categorisation of transactions. This intelligence creates cross-selling opportunities, as financial institutions and Fintechs realise the most suitable services for users based on personalised financial transactions’ analysis.
Fraud prevention: Mining data from user's transactions can determine normal and abnormal behaviour. AI can build up individual profiles exposing anomalies such as unusual transactions. AI can also thwart cyberattacks by continuously learning from interactions with analysts and customers to deliver scaled-up intuitive intelligence. AI technology for detecting telephone fraud is also in place, analysing suspicious calls by factoring in geo-location, call type and unique caller identity. This technology is currently deployed in the US for banks and retailers, and it’s only a matter of time until every formal telephone interaction is filtered through anti-fraud software.
Advice: Robo Advisors are online wealth management services providing automated, algorithm based portfolio management or investment recommendations. It is estimated that Robo Advisors will handle $.2.2 Trillion by 2020. Although Robo Advisors currently operate in the ‘passive asset’ class, they may yet become a disrupting force in asset allocation, driven by technological advances and low barriers to entry.
Customer service: In 2015 the Mizuho Financial Group Inc Bank in Japan sent Pepper the humanoid Robot into its Tokyo branch to handle customer enquiries. A partnership with IBM will enable Pepper to be built into apps to analyze data, make recommendations, and even understand human emotions. Interesting though this is, the most popular form of bank-bot will be online automated research and customer service assistants. An example of this is RBS trialing “Luvo” AI customer service assistance to interact with staff and customers.
Smart insurers: Another development is the use of bots to engage with customers on a personal level, as is seen in the insurance sector. AXA have an app based bot called Xtra by AXA. Based on data gathered through wearable devices and mobile apps, the bot called ‘Alex’ engages in bespoke conversations with AXA customers about healthier living. This dialogue facilitates insights and personalised rewards and encourages customer engagement. Based on the consumer’s lifestyle knowledge, the smart insurer will be able to adapt and personalise its premium rate.
Lending: Data-driven AI applications will speed up online lending. Online brokers, lenders and banks can use algorithms to assess eligibility for credit, and AI can also match business owners with the right lender; for example, connecting entrepreneurs with the best lender for peer-to-peer lending. AI analyses and authenticates user’s transactional data, and income verification and spend analysis helps highlight risk factors used for a richer credit scoring experience. This will reduce the risk of default and increase borrowers' financial knowledge.
And finally… AI expert and futurist Ray Kuzweil predicted the internet’s ubiquity and the rise of mobile devices in his book The Age of the IntelligentMachines (published in 1990) and has also said AI will exceed human Intelligence in 2029. By 2029 we’ll be ready, and while as in the case that customer service bots can emulate advice, our ultimate preference will nearly always be for human contact, even arguably amoung tech-savvy millennials. As AI continues to a resemble a Hybrid Intelligence framework – a system leveraging what humans do best (imagination) and what computers do best (calculation), we have much to look forward to.