How AI could transform consumer lending
Widespread deployment of AI/ML throughout an organisation, well beyond the traditional implementations in CX or underwriting, has the potential to transform consumer lending through the elimination of inefficiencies, waste, and pass-along costs, argues Tamir Hazan.
Whether the average consumer is aware of it or not, artificial intelligence (AI) and machine learning (ML) are active somewhere in the background of almost every tech-based interaction, from ordering on Amazon to making an ATM withdrawal, to applying for a mortgage. While AI and ML are both linked to exponential growth in the fintech industry, for expansion to be sustainable and responsible, fintech must look deeper than yesterday’s AI’s applications and consider a reboot.
AI/ML 1.0 limits itself to yesterday’s uses. Its scope tends to be narrow and focused on distinct fintech functions, such as underwriting, chatbots, and other similar use cases. While those functions remain integral to today’s market and consumer-driven interfaces, the promise of AI/ML in the fintech ecosystem, and the expected cost savings and revenue enhancements, lies in integrating AI/ML into every stage of business, both on the surface and behind the scenes.
In 2019, IDC projected that banking will be the second largest global industry to invest in AI. There is wide consensus in fintech that the future of the industry depends on effective AI/ML integration. There is significant opportunity beyond those businesses that focused on a small slice of AI/ML functionality and stand to reap only a miniature portion of the potential return.
For instance, while superficially the digital lenders have reported incredible top line origination numbers (digital lenders report growth in $50 billion per year), AI/ML-driven digitization of the lending process hasn’t broadly been able to overcome high marketing costs and achieve profitability. This has led some to question how much additional growth is possible. As this moment, fintech disrupters are beginning to show that as the focus in AI/ML shifts away from a few specific uses to maximizing potential by addressing systemic inefficiencies and achieving cross-organizational integration, further growth is possible.
What’s at Stake and Potential Roadblocks
AI growth is trending and seemingly unstoppable. Accenture has predicted AI could lead to an economic increase of $14trn by 2035.
“But before financial institutions can reap all of AI’s benefits, they must first overcome challenges, including security, privacy, bias, and regulatory issues.”
Every millisecond, there is new data to work with. Everything from credit history, to marital status and whether an individual is a Mac or PC user, is traceable. How can companies responsibly gather, interpret, and act on data they collect?
Sometimes what data they are collecting is as important as how they choose to manage it, maybe more so. Furthermore, limited, or skewed, data can lead to biased algorithms, a net result of replacing human bias with machine bias. A lose-lose situation.
As AI/ML knowledge and use applications increase, so do consumer and societal concerns about confidentiality, the potential for bias and/or discrimination, and transparency. Fintech companies must focus on sound, long-term growth, along with customer relations, as they amass data and decide how to manage it. Operating from a profit-at-any-cost mentality is shortsighted and will not stand the test of time.
Likewise, those companies focused on AI/ML 1.0 uses, ignoring the potential for wider and more integrated AI/ML applications, will demonstrate slowing to stagnant growth.
AI/ML 2.0: The Future of Fintech
Businesses, including fintech disruptors, that are ready to embrace innovation through novel AI/ML applications and those who consider company-wide integration of AI are those best poised to take the lead in the field and achieve accelerating growth profits.
Thoughtful planning, management, and integration of AI across organizations all lead to greater returns, part of which can be passed back to customers, a win-win scenario.
AI/ML 2.0 requires new ways of looking at old business models. For example, applying AI/ML by creating an algorithm for a collections call centre might lead to more calls that generate payment. This allows businesses to more thoughtfully allocate resources; time gained (less “wasted” time on calls that don’t yield results) equals profit, along with resources and attention to direct elsewhere, boosting business and the bottom line.
AI/ML 2.0 also needs AI experts and data analysts to lead the way. Innovation will depend on analysts and AI/ML specialists to make sense of the rapidly emerging field of AI and its seemingly infinite potential applications. Even as AI may reduce the human workforce in some capacities, the need for deep AI/ML understanding and data mastery will be requisite skills for newly created jobs.
To date, some level of inefficiency is considered a usual part of doing business. That no longer needs to be the standard.
With the help of AI/ML 2.0 and business leaders who recognize its cross-functional potential for streamlining processes, increasing efficiency, and yielding growth, the future of fintech looks bright.