As artificial intelligence (AI) gets smarter — as it’s designed to do over time and as usage increases — there are more places in the payment and connected economy ecosystems where it can have profound impacts from payments to fraud mitigation to customer service.
A shrewd observer of market trendlines, Edwards sees huge opportunities to bring AI into areas like mobility, particularly as it relates to the connected car and the payments opportunities it can advance for consumer convenience on one hand, and sales on the other.
“Apple CarPlay and Google [Android Auto] have taken most functions away from the driver and turned them into voice commands. There has got to be some element of intelligence built into that because there’s a dialogue that goes on,” he said.
“But it could get so much better if it was tied into my car and [saying] hey Drew, you’re down to an eighth of a tank of gas and there’s a QuikTrip up here on the right which already has your payment credentials.”
Ditto for EV drivers looking for a nearby and available charging station with no wait time.
“It’s totally transformative, and payments are in the middle of that,” he said. “It’s where the connected car comes in because you can’t use your hands, and you’re interacting with this artificial intelligence. All the way through to paying for that experience, it just happens without you ever touching a keyboard or putting your credentials in. I love that.”
Taking these uses a step further — which is essentially the point of AI — Edwards is fascinated with how Waze serves ads while mapping a drive, although he feels this still needs improvement, which will happen as systems “learn” from each interaction and transaction.
Instead of pinning a McDonald’s ad to a Waze map — which can be difficult to act on while driving — he said, “The advertiser could impact the recommendation that was made instead of the QuikTrip up here on the right, there’s a RaceTrac up here on the left, why don’t you go there? And by the way, they’re 5 cents a gallon less. Beautiful experience.”
While AI is widely deployed in financial services already, it’s just scratched the surface of what’s possible there.
While he poked good-natured fun at how Siri and Alexa still give some zany answers to queries, he does see a place for voice AI in financial services. He’s even picked some early favorites.
“I have been somewhat impressed with Bank of America’s Erica®,” he said. “There’s some AI built in there and it is getting better and better. When I ask a question, simple things like show me all the bills I paid last month, or show me all the recurring bills, or show me all the utility bills, it can find all of that. That’s the beginning of really improving a customer experience.”
Customer experience is another greenfield opportunity for AI. For example, he sees the somewhat irritating chatbots of today as moving towards something transformative.
Applied to areas like eCommerce, it’s already altered the online experience dramatically.
Speaking to Amazon’s use of AI particularly, Edwards said, “They have taken a very complicated shopping experience, and using a lot of that tech, boiled it down to a customized shopping experience for each Amazon shopper. If we can just apply that to the pay-in and payout experience, to the checkout experience,” it would be an experiential sea change.
Along similar lines he noted that Ingo processes $9 billion a year in digital payments from dozens of sources, saying, “We’re beginning to look at where we can use AI to help with that reconciliation process, because you’re looking for anomalies. That’s where you’ll spot fraud or something wrong. I can’t think of a bigger place than payments where AI can improve the experience.”
There are barriers to overcome, however: “True AI is not really in the grasp of most companies right now, because that takes scale,” he said.
From customer service to back-office automation, he feels there’s lots of room for AI to improve “around the payments processing experience. There are still too many people and too many reports and too much reconciliation going on that could be done by a machine.”
“I don’t think you eliminate people,” he said, but “you can get much better at sorting down to just the things that humans need to make a judgment call on using AI. That’s what we do.”
AI’s ability to sort through vast amounts of transactions and detect anomalies goes back to the core fraud mitigation use case that most financial institutions and merchants are already using.
Saying two-factor authentication “is kind of coming to the end of its cycle as the crooks have figured out how to spoof it and beat it,” AI can vastly modernize know-your-customer (KYC) and identity verification (IDV), for example.
Harking back to voice AI use cases, he said “without a doubt, I think we are going to get away from the keyboard. Then when you apply biometrics, including voice prints, with AI, if I’m buying with my voice, they literally should be able to tell if it’s me or not.
“From onboarding and the opening of new bank accounts and even the security to log in, there’s a huge place for AI in improving customer experience,” he said.