By Adam Atlas
Attorney at Law
If we believe the promise of tech leaders like Elon Musk, in a few short years, we will live in world with self-driving cars, robots and a multitude of services powered by artificial intelligence (AI). If you don't believe we are headed for that future, skip this article; it won't be worth your time.
Thank you for reading on. As objects such as cars, shopping carts, delivery robots, refrigerators, drones, surveillance cameras, phones and countless other things around us, including our pets, become connected to what will be a 5G cellular network moving 10 times more data than the current 4G network, it makes sense to query what will be going in payments from a legal perspective.
A self-driving electric car that needs to recharge and does so automatically at a charging station will have to make a payment for the electricity. The owner of the car will have to load the car's wallet with a payment instrument that can be presented to the charging station.
Yes, the car and the charging station can be part of a pre-programmed network of charging stations and cars that use them, but if they are not pre-programmed, they need to have a common form of payment.
This provides an example of machines making payment decisions without direct human involvement. Presumably, the car owner will give the car instructions and permission to charge up where necessary and make the corresponding payment. In any case, what is important is that humans will be, at last, one step removed from some future payment scenarios.
Imagine being accustomed to seeing lawn-mowing robots prowl your neighborhood. And imagine you can pre-program your home to accept having the grass cut by any robotic lawnmower that is willing to do the job for $7.50. Indeed, we might see a time in the near future when owners of mower robots will set their offer prices, and homeowners will set their offer prices, and an automatic search for matches will occur as robots canvass the neighborhood. Where prices agree, or where the home's AI and robot are authorized to negotiate within preset parameters, transactions will occur.
Today, most consumer payments are made at the moment when the consumer has all the relevant information about the purchase. The gardener is on your doorstep and offering to cut the grass for a known price. In the automated home and robot mower scenario, payments decisions are made a bit like stock market purchase decisions – where we input a set of rules that – when triggered – will initiate a payment.
Legally, beneficiaries of these payments will have to prove that the consumer who authorized them did so knowingly. We can foresee some confusion on the part of consumers about the various payments they pre-programmed, and they might then claim not to have understood the implications of their programming decisions.
We see this today in subscription sales and future-delivery sales – areas of commerce that typically experience high levels of chargebacks. To avoid this, consumers may have to improve their memories or merchants and robot programmers may have to come up with novel ways to explain and obtain payment consent rules from consumers.
Automation and AI will make some activity more predictable. For example, based on our driving habits, electric self-driving cars could be preprogrammed to adopt their owners' best habits and use them without fail, for example, always signaling a certain number of feet before a changing lanes.
I prefer not to predict what kind of new payment methods will be available. Instead, I'll envision making use of today's known payment methods but wire them into future automated robot payment scenarios.
For a number of years, federal banking regulators have been working on faster payments, and it is safe to imagine that instant value transfers will be possible on conventional payment rails (not bitcoin) within a few years. Consequently, the payment methods are likely to be instant automated clearing house, credit card, debit card or stored-value spend – each using rails that are familiar to today's processors.
On the issuing side, banks will need to enable APIs to connect their payment cards to robotic and automated devices – like self-driving cars. This is relatively simple process and does not require much more technology than already exists today.
On the acquiring side, there will have to be rules of play, like use of contactless payments, to enable acceptance of cards and other payment methods in a manner that is known to both paying robots and payee robots.
Again, the existing acquiring processors and economic assumptions do not need to be revised to function in the service of automated merchants. Today's vending machines already do this when you purchase a pair of headphones at the airport by presenting your credit card to the machine. Where the payor is automated, we simply replace the human cardholder with a machine.
There are three primary considerations for payments in an automated world. First is consent. Consumer consent is hard enough to prove now (as discussed above) and will be even harder to prove in cases where consent becomes a consent not for a single payment but for a complicated set of rules involving your home, grocery store, car, pets, family members etc.
The rules will be so complicated that typical consumers might not want to be bothered to understand them all – and there might be underwriters who will enter the market to insure the risk of faulty automated purchases. In today's world, a three year-old-child plays with an iPad and spends $1,000 on game tokens. In the automated scenario, your refrigerator orders a truckload of milk by accident.
A second consideration is privacy. It sounds old fashioned to talk about privacy when we have so little of it left – but I believe legislators will step up to provide some rudimentary privacy laws, such as the new California Consumer Privacy Act.
Finally, automated transactions in the physical world are a slightly new area that is yet to be built. Contracts of purchase and sale initiated and executed by machines require drafting and understanding legal terms that are more complicated than the current purchase and sale terms. These terms will challenge lawyers – or their AI substitutes.
It is worthwhile for ISOs to learn about new automated machines to seize the tremendous opportunities that they present for payments we have not yet imagined.
In publishing The Green Sheet, neither the author nor the publisher is engaged in rendering legal, accounting or other professional services. If legal advice or other expert assistance is required, the services of a competent professional should be sought. For further information on this article, please contact Adam Atlas, Attorney at Law by email at email@example.com or by phone at 514-842-0886.
Notice to readers: These are archived articles. Contact names or information may be out of date. We regret any inconvenience.Prev Next