A recent Financial Times article by journalist Dave Lee discusses the tension between humans and the “drive” towards autonomous vehicles. “Buckle up. We’ve hit peak Human-Autonomy Clash.” It is a fascinating read that weighs the safety concerns of real-world road testing, and the tactics employed to help us mere humans become comfortable with the idea of autonomous cars (and buses, and trucks) sharing the road with, well, all of us.
Perhaps a substory here is how reliant autonomous cars are on data, lots of it. A vast amount of data has been harvested for years and is ingested into sophisticated AI models to instruct the vehicle in real time. Leading such initiatives also requires a high degree of confidence and trust in data, and the science applied to it. Successful model performance (in addition to other factors) creates reliable outcomes – safe journeys.
What, you might ask, does this have to do with banking, and corporate banking in particular? Banks also have masses of data, but it is often viewed merely as a byproduct of banking operations. Increasingly, data needs to be viewed as the foundation for banking services in an information-based economy. As the role of the treasurer becomes more strategic, treasurers now expect banks to deliver insights and analytics-based solutions to help them manage their business. This gives rise to the “intelligent treasury,” where analytics and insights on top of quality data are embedded in client-facing solutions. Perhaps the future will be an autonomous treasury where only exceptions require intervention.
Just as the implementation of AI to drive cars autonomously wasn’t introduced overnight, neither will the autonomous treasury (or bank for that matter). AI has been applied incrementally into vehicles for several years now as driver aids. Blind spot indicators, lane departure warnings, steering “correction,” and accident avoidance technologies are but a few.
Leading banks take similar incremental approaches. In delivering the intelligent treasury today, I prefer to use the term, “augmented intelligence.” As with driver aids, this is an approach where data science is paired with human interaction to help enable learning and enhance decision-making. Think about the “treasury aids” of cash forecasting, risk modeling, and payment optimization as examples.
Whether driving aids or treasury aids, some core data characteristics are essential. Knowledge about the data, expertise in managing data, and trusting the data for critical decisions. Before delivering the intelligent treasury, banks and treasuries need confidence in their data, data talent, and a data-centric culture for solution development.
As we come into conference season, I’ll be interested to see how banks are progressing toward intelligent treasury. If you are struggling to find direction going forward, check out recent Celent Celent Corporate Banking Research into data strategy, intelligent treasury, and adoption of AI.
Perhaps the future will be “self-driving” treasury services. Until then, keep your eyes on the road!