By Bradley Leimer, Managing Director and Head of Fintech Strategy, EAC
The concepts fueling the development of artificial-intelligence applications are hardly new. Attempts to understand how the brain thinks—how decisions are made, how it saves and processes memories, and how it leverages connected learning systems—have been perplexing humans since before the times of the classic philosophers. Soon after brilliant English mathematician Alan Turing published his groundbreaking 1950 paper “Computing Machinery and Intelligence”, the academic discipline aimed toward one day surpassing human intelligence was born.
The term artificial intelligence (AI) itself was coined by Boston-born computer scientist John McCarthy in 1956, as he and his peers established research labs at MIT (Massachusetts Institute of Technology) and Stanford. The current wave of AI startups owes much to the decades of research and experimentation developed by these teams and those that followed. As it pertains to financial services and the creation of new sources of efficiency and customer value, however, the application of cognitive computing, machine learning and AI is, in many ways, a greenfield.
Global financial-services firms are currently undergoing costly digital transformations as they innovate a centuries-old service model. Existing providers are being challenged by external venture investments threatening their customer relationships and revenue streams. While banking has a history of resisting or delaying deployment of modern methodologies—agile development, open source, cloud computing, platforms—leveraging artificial intelligence is one area that the industry appears to be readily embracing.
The twin catalysts: investment and innovation
The influx of new firms focused on AI is simply staggering. There are now more than 2,300 AI startups globally, half of which did not exist just two years prior. In regard to investment, the number of venture transactions funding AI startups increased nearly 10 times over the last six years, from 67 in 2011 to 658 in 2016. Last year, startups leveraging AI as a central component of their business raised a little over $5 billion, a number that is expected to double to more than $10 billion this year (CB Insights). Technology giants such as GAFA (Google, Apple, Facebook, Amazon) are investing tens of billions more. Intense competition for people, products and patents has erupted between the United States, China and other tech centers. Research firm Tractica forecasts that revenue generated from AI solutions will grow from $1.4 billion last year to $59.8 billion by 2025. Given the rapid growth in the space, the numbers are likely to be much higher.
While not all of these companies are focused on financial services (we still need to improve self-driving cars, parcel shipments and Instagram selfies, after all), banks are increasingly looking at AI startups for sources of inspiration. One global bank that clearly sees the promise of AI and machine learning is the Spanish giant Banco Santander. InnoVentures, Santander’s $200 million venture fund (now the most active bank-backed fintech investor), recently invested in three AI-focused startups. Pixoneye leverages stored photos to build customer profiles and unearth customer needs (photos of travel destinations might get you a travel-partner discount, for example). Gridspace analyzes voice patterns in real-time and determines the mood of the customer during service calls to adjust agent scripts accordingly. Curve links customers’ debit and credit cards to a single app and a Curve-branded MasterCard to optimize payments and better manage everyday spending. InnoVentures also invested in Elliptic, Socure and Personetics, all of which leverage customer data and machine learning. Adding more intelligent experiences into its digital-application stack will provide ways to engage its customers and differentiate its service offerings.
Other banks have made notable investments across fintech—and like Santander, increasingly in AI. The most active European-based bank-led corporate investors are Santander, UBS, Deutsche Bank, Société Générale, BNP Paribas, Credit Suisse, HSBC, BBVA (Banco Bilbao Vizcaya Argentaria), Barclays, ING, UniCredit, RBS (Royal Bank of Scotland) and Crédit Agricole. Interestingly, 70 percent of investments from these European banks went to US-based fintechs. The top US-based bank investors include Citi, Goldman Sachs, JPMorgan Chase, Morgan Stanley, Wells Fargo, Bank of America, TD Bank, Capital One, US Bank and PNC. These American banks have participated in 72 rounds totaling $3.6 billion to 56 fintech companies.
Clearly, these institutions are focused on combatting external threats and their evolving business models—but investment alone does not equal innovation. True innovation involves a long-term focus to apply these investments, partnerships and rapid internal developments toward services your customers love and are willing to pay for.
Operational efficiency vs. customer centricity
Artificial intelligence and the umbrella of associated cognitive tools have proven to be a catalyst for both business-model reinvention and increased efficiencies. Advanced tools are casting a wide net of influence on financial-application development. These include improvements in KYC (know your customer), regulatory and compliance, portfolio management, algorithmic trading, security, collections and fraud detection, insurance and credit underwriting, customer service and engagement, predictive analytics, and product fulfillment.
Applications leveraging machine learning, deep learning, pattern recognition and natural-language processing can work together to simulate human-thought processes. With the rapid consumerization of artificial intelligence—seen through the rise of contextual digital experiences and voice-driven intelligent-response platforms, the way customers interact with financial brands is expected to shift significantly. Therein lies the rub.
Should management teams focus on customer-facing improvements, or on internal efficiencies that are often invisible? While much of the current wave is focused on operational efficiencies, the most strategic executive teams are firmly focused on creating new forms of customer value and revolving revenue streams. Here is a selection of efforts that demonstrate the principles of optimization, and to a lesser extent, customer centricity.
JPMorgan Chase is leveraging AI to reduce costs and reduce legal research time by thousands of hours. The initial project, called COIN, for Contract Intelligence, interprets commercial-loan agreements and should save 360,000 hours of work each year by lawyers and loan officers. The software reviews documents in seconds instead of hours, and runs 24/7. Chase also developed a proprietary AI-based trading platform (LOXM), which was trained on billions of historic transactions to execute equities trades at maximum speed at optimal prices. With Chase’s $9.6 billion annual technology budget and digital developments, it is one bank to watch for further developments in the AI space.
HSBC is undergoing a technology renaissance of sorts. It is incorporating technology from Ayasdi to tackle fraudulent activities such as money laundering and KYC. HSBC is focused on gaining efficiencies from automating transactional-data analysis, and like several other banking entities, it seeks to gain a significant reduction in cost by eliminating false positives in fraud detection. It has already seen a 20-percent efficiency gain in initial pilots.
Citi has invested in Feedzai to detect and deter fraud within transactional applications. Like HSBC, Citi is working with Ayasdi (and Cylance) to learn how to leverage machine learning across its international networks. Citi has invested in Clarity Money, which leverages machine learning to get customers to best allocate their non-allocated funds to earn better returns. With its focus on investment and partnership in the fintech space, Citi is a firm that will continue to demonstrate significant innovation in this area.
Morgan Stanley is moving beyond the concept of robo-advisor to provide better tailored advice. It is investing in a platform to make its financial advisors more efficient. Advisors will offer a slew of passive-investment options and ETFs (exchange-traded funds) but can also provide tailored investment options based on firm research and client preference. The bank is also investing in personalized portfolio alerts, which could include low-cash-balance alerts, margin calls or real-time changes in a client’s portfolio. These alerts may also leverage social sentiment, which can impact market movements and short-term portfolio value. Most interestingly, its new system will include content surrounding life events—a marriage, the birth of a child, particular medical needs, job relocation or other events that create the need for particular financial strategies for that event. Behind the scenes, this is all powered by AI.
LPL, the largest independent broker-dealer, also thinks its advisors can leverage AI to enhance client relationships. Rather than looking to Betterment or Wealthfront to simply reduce investment-management fees, it is looking at Acorns and Trim for its inspiration. By continually analyzing personal accounts each day, it hopes to find savings opportunities and more efficient ways to spend. Its staff leverages machine learning to help understand micro payments on the aggregate level to reduce long-term drag on investors’ current accounts.
Wells Fargo founded a new team focused specifically on AI in early 2017, led by long-time innovation head Steve Ellis. Projects range from applications to identify payment fraud and employee misconduct. Staff will also develop technology that can better create personal recommendations to customers on a variety of financial products. Wells Fargo is working to enhance its automated client communication through Facebook Messenger, and continues to expand services offered through its vast ATM fleet. All of these services are automated.
AI-driven applications are being added to chatbots, robots and physical locations. India-based City Union Bank customers can interact with its robot named Lakshmi that can respond with basic account details. Mitsubishi created a robot, Nao, that analyzes facial expressions and behavior and can talk to clients in multiple languages. HSBC launched several learning chatbots (Andrew, Amy and Olivia) to answer client questions from various segments within both the bank’s internal and external social applications, such as Facebook. RBS leverages Luvo to help customers with simple financial tasks. Capital One and American Express customers can check their accounts and pay credit-card bills through Amazon’s Alexa. And there is apparently much more to Alexa than a simplified robotic voice. USAA, long known as an innovator, has become the most vocal actor in the AI theater.
USAA: a trailblazer in the AI space
USAA has been experimenting with AI and machine learning for some time. With its new Alexa app, USAA worked with Clinc to create a conversational interface that members can use to more fully understand their finances. Rather than requiring scripted requests about balances, USAA members can ask Alexa questions in more natural sentences, such as “Can I afford the new iPhone?” or “How much did I spend on that business trip to New York last September?” The key to what USAA has built is that it is experimenting with conversations around its members’ money, not just a vocal version of online or mobile banking. It is clear its strategies revolve around the needs of members, not an afterthought as it often feels like at other financial entities.
How can the USAA Alexa app and similar efforts enhance day-to-day financial activities? The difference between algorithmic applications today and the learning systems of tomorrow will be time horizon and breadth of advice. Within the AI framework, it becomes possible to provide a hyper-personalized, fully integrated, contextual relationship with each customer. Financial brands can then demonstrate a deeper understanding of individual constraints and opportunities. How well we show empathy around the context of our customer becomes paramount. Rather than requesting information from the application, it will truly become an ongoing conversation about broader topics—whether they have immediate or much longer-term financial consequences.
Everyday decisions, optimized toward a desired end state
While you might look at artificial intelligence as simply a different way of developing applications, as computational capabilities increase, we will see a proliferation of life-enhancing applications. These will likely have significant societal impacts. Like USAA’s initial efforts with Alexa, we can create a deeper sense of intimacy that wasn’t possible even a year ago. Imagine where we will be a decade from now. AI reinvigorates possibilities driven by customer centricity and by leveraging the data of the individual over time. It is no longer just current context, but an accurate roadmap to a desired future state. It becomes a multi-generational approach to building customer value—incremental steps toward further stabilizing wealth and equality by providing new levels of financial opportunity.
Helping manage your finances should be much broader than numbers in a bank account. One way to view financial placement within a given society is opportunity meeting action. While day-to-day financial activities such as saving and investing matter over time, a great deal of life’s opportunities are often predetermined. The Horatio Alger story is a nice tale, but the conditions still have to be right for a rags-to-riches story without a lot of help, luck or both. Broad financial developments enhanced with AI will help change that equation—to help people make the right decisions at the right time—or more importantly, over time.
As financial applications extend their breadth of advice and learn from billions of potential paths, we can imagine assistance across every type of life event. Where you choose to live or visit (if you have options for mobility, how frequently you travel); What educational opportunities you pursue for yourself and your children; What type of health care is affordably available (your life’s duration, health and related financial implications); What line of work you choose (and how you take opportunities to improve your financial position through your career); Are you able to start a business? How do you spend your leisure time (and subsequently your money)? How are you accumulating wealth, and how will you decumulate wealth and extend it to subsequent generations? The breadth of advice will include micro-decisions that lead up to these larger critical paths.
Developing a new social contract.
As artificial intelligence helps usher in the fourth industrial age, how will banks and AI-embedded financial applications impact the greater society? How could financial players create a new social contract? It is possible that new forms of smart applications can make society more equal? Similar to the eradication of disease through persistent inoculation, we may be able to finally limit the impact of systematic societal issues such as poverty and income inequality. Over several generations, universal access to smart financial applications will make society stronger—by levelling up personal finances and associated opportunities to access education.
Rapid technological advances have only begun their impacts on humanity. Financial institutions act as both gatekeepers and enablers and have a huge role to play. While financial executives may not subscribe to the dire mantra of innovate or die, their directives to their teams and their words to shareholders matter. Publicly working on infrastructure and efficiency projects to reduce costs may be signals to the equity markets that your profits will increase; but like corporate restructuring, this demonstrates a lack of imagination and a short-term view. It is time to evaluate how the future of financial applications will benefit not just the bottom line but global society.
What is your financial-services firm doing to be part of the future?
Additional Resources History of artificial intelligence (Wikipedia) - https://en.wikipedia.org/wiki/History_of_artificial_intelligence John McCarthy, 84, Dies; Computer Design Pioneer - http://www.nytimes.com/2011/10/26/science/26mccarthy.html?hpw Artificial Intelligence Software Revenue to Reach $59.8 Billion Worldwide by 2025 - https://www.tractica.com/newsroom/press-releases/artificial-intelligence-software-revenue-to-reach-59-8-billion-worldwide-by-2025/?utm_content=buffere05a0&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer For AI startups, more funding is often not the answer - https://venturebeat.com/2017/07/01/for-ai-startups-more-funding-is-often-not-the-answer/?utm_content=buffera5cf7&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer What Santander’s latest bets say about the future of fintech - https://www.americanbanker.com/news/what-santanders-latest-bets-say-about-the-future-of-fintech Santander Buys Stakes in Three Startups as Botin Pushes Into Fintech - https://www.bloomberg.com/news/articles/2017-07-12/santander-buys-stakes-in-3-startups-as-botin-pushes-into-fintech Robo-Advisor Betterment Is On A Personalization Push As It Surpasses $10 Billion In AUM - https://www.fastcompany.com/40442080/robo-advisor-betterment-is-on-a-personalization-push-as-it-surpasses-10-billion-in-aum HSBC adopts AI to streamline compliance efforts - https://banknxt.com/61290/hsbc-ayasdi-ai/ Banking Must Move From Mobile-First to AI-First - https://thefinancialbrand.com/65338/banking-ai-ui-artificial-intelligence-data/ Rise of Artificial Intelligence in Banking - http://bfsi.eletsonline.com/rise-of-artificial-intelligence-in-banking/?utm_content=bufferee4bb&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer The double-edged sword of Artificial Intelligence - https://www.linkedin.com/pulse/double-edged-sword-artificial-intelligence-brett-king How Artificial Intelligence Will Revolutionize Banking - http://www.informationweek.com/big-data/how-artificial-intelligence-will-revolutionize-banking/a/d-id/1329218 Machine Learning in Finance – Present and Future Applications - https://www.techemergence.com/machine-learning-in-finance-applications/ How Machine Learning Is Helping Morgan Stanley Better Understand Client Needs - https://hbr.org/2017/08/how-machine-learning-is-helping-morgan-stanley-better-understand-client-needs?utm_campaign=hbr&utm_source=linkedin&utm_medium=social How microlending could end extreme poverty - https://www.weforum.org/agenda/2017/06/how-microlending-could-end-extreme-poverty?utm_content=buffer84cb9&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer How Artificial Intelligence Could Democratize Financial Services In Asia - https://www.forbes.com/sites/chynes/2017/07/03/how-ai-could-democratize-financial-services-in-asia/#495ba5e6618a How Artificial Intelligence Will Revolutionize Banking - https://aitrends.com/financial-services/how-artificial-intelligence-will-revolutionize-banking/ Artificial Intelligence In Financial Services – Resistance Is Futile - http://irishtechnews.ie/artificial-intelligence-in-financial-services-resistance-is-futile/ What will we do when we don’t need to work? - https://thefinanser.com/2017/08/will-dont-need-work.html/ What you should know about AI - https://techcrunch.com/2017/08/01/what-you-should-know-about-ai/
Very well written article. I hope AI improves & builds on data quickly, very much looking forward to people having th technology to simplify their financial goals.
This post is worth everyone’s attention. Where can I find out more?