Jaron Lanier at the Tribeca Film Center on April 23, 2018 in New York. (Photo by Ilya S. … [+]
I was interested to see that Gartner
HE
APIs, not brands
This all seems a bit hyperbolic, but the fact is that I think Gartner might just be right. They suggest that organizations pay strategic attention to the shift from human to machine clients, and I have long believed that this should be a priority for financial services organizations in particular.
Since I know nothing about marketing, I’m fascinated to see how banks will adapt acquire robot customers who don’t care about the bank’s logo, TV commercials, or the sports team it sponsors. So when my smart wallet uses open banking data and decides I need to open a savings account, get a loan, or refinance my mortgage, how will my financial robot decide which provider to use? After all, I don’t really want to know because I have better things to do.
As I’m sure is true for most people, most of the time, in the not too distant future, our financial decisions, transactions, and analyzes will be carried out by robots operating within relevant due diligence legislation with the coordinated aim of ensuring financial health. I don’t think that’s a bad thing, because I’m sure that even a rudimentary financial robot can do better than me when it comes to money management.
Since I intend to give responsibility to my financial robot. so, how will this robot choose which accounts to open, which services to use and which oracles to listen to? I imagine it will use a combination of reputation and other relevant data (e.g. economic forecasts) to determine which account is currently there, then I’ll just click OK and it’ll be done. Calculating reputation will of course involve fees and rates, but instead of using the Victorian brand substitute for real data, my bot will look at API functionality, open financial interface availability, service availability, etc.
(I say a “Victorian” substitute because the first registered trademark in the United States, filed for paintings, was issued on October 23, 1870 and the first registered trademark in the United Kingdom was issued on January 1, 1876 for the triangle red from the Bass brewery.)
This means that banks, financial organizations in general and, of course, fintechs will sell their products to machines and not people. Well, their machines will sell things to customers’ machines. Now, people have tried letting AI make financial decisions in the past and, truth be told, it hasn’t worked very well. AIEQ of the ETF Managers Group
AIEQ
IBM
How did that happen ? Well, over the last five years, it returned 4.9%— behind the 11.78% five-year return of Vanguard’s benchmark S&P 500 index fund and, for another comparison, two large actively managed funds (the American Funds Growth Fund of America, at 9.81%, and Fidelity’s Contrafund, at 11.04%). This doesn’t seem particularly successful to an amateur observer like me.
Yet many people are foolish because they think that historic robo-advising was essentially modernized machine learning. The customers Gartner is talking about will use AI and deep learning algorithms to deliver something very different and will require very different services from financial institutions. As an obvious example, businesses may need to provide specific APIs to meet the needs of bots rather than people, as bots can search through more data, access more sources, and process more transactions than n any person. Service levels acceptable to a customer may be completely unacceptable to a customer.
Captchad!
These customers may not be that far away. The Commonwealth Bank of Australia (CBA) is already exploring how it can use generative artificial intelligence create fake consumers who can test new products. Dan Jermyn, their chief decision scientist, said the technology can allow machines to perform experiments on products to determine their popularity. They leverage simulated daily life experiences to mimic behaviors to improve qualitative and quantitative understanding of how customers might respond to changing contexts, daily financial challenges and new products.
It’s a pretty fun way to create a SimBank, but it’s surely not a big step in turning these client bots into client bots. If you know what I mean.
Money machine
By the way, bot-to-bot transactions could take us into new territories when it comes to money. The German Banking Industry Committee (GBIC) is the voice of the leading associations in the German banking industry. I had always thought they were a very conservative organization, so I was fascinated to see that they call for some form of private sector token money be developed to meet business demand arising from Industry 4.0 and the Internet of Things. They envision that such money would facilitate transactions based on “smart” (i.e. automated) contracts and thus increase the efficiency of processes.
(The idea of giving machines the money they need to spend may seem pretty radical, but Commerzbank was already testing blockchain-based machine-to-machine payments between electric charging points and Daimler Trucks in 2019.)
Gartner is surely right to see this bot-on-bot action as a priority for financial services in the future and these transactions will undoubtedly lead to astonishing changes in financial services. But even Gartner may be underestimating the changes they will bring to money itself.