Over the next decade emerging technologies and an increasing human-machine partnership will impact virtually every aspect of our lives. The finance sector may be particularly open to increased human-machine partnerships. Already, machines drive most of the transactions in stock markets. Moving forward, the emergence of technologies like distributed-ledger blockchain systems, machine intelligence, and automated financial purchasing promise to further transform this space.
So, what does the next era of human-machine teams mean for finance by 2030? How might these emerging technologies influence the way we track financial information and account for resources? Can the fusion of human and machine intelligence help to increase the security of our financial infrastructure?
To explore the opportunities, Institute for the Future published a foresight vignette that offers a first-person view of how the next era of human-machine collaborations and co-dependencies may reshape how we administer financial system. Metromile, a car insurance startup that offers pay-per-mile insurance and a smart driving app, is highlighted as a leader in the field. The following is an interview with Dan Preston, CEO, on how financial sector will change over the next decade.
Q: How plausible do you find the scenario, Dispatches from a Bot Operator?
A: The concept of a service or an algorithm acting on the customer’s behalf to find the kinds of products that fit that person’s life during any given time is a very likely outcome. I’m not even sure it’ll take 10 years for that to come to fruition.
We will have more and more products that leverage data to give consumers an exposure-based solution—insuring events as they happen as opposed to offering only blanket policies. The Metromile model is a first version of that, but this concept will start to impact a lot of other types of insurance.
Q: What would accelerate this future?
A: The examples of a self-driving car automatically finding optimal coverage for itself and the home security system automatically increasing coverage are pretty interesting. The technology is there to make that happen so I fully expect that it will be possible over the next ten years. The question is how will customers adopt it.
My guess is that people won’t be willing to have these variable costs associated with services. Instead, customers will buy services at a flat rate from providers and the providers will incorporate exposure-based insurance models into the cost of their services. So, for your self-driving car, for instance, you wouldn’t see the variable cost of car insurance. It would be built into the pricing model of the car. I think this is because there is a requirement for transparent pricing and little fluctuation, especially for situations in which consumers have little control.
Q: What would prevent this scenario from happening?
A: The biggest friction is getting access to the data. Rightfully so, people are reluctant to share their data with a broad number of companies. You have to offer value for that data at the very beginning. But, often times, in order to create that value, you need the data in the first place. It’s a bit of a challenge. Metromile was fortunate because customers were willing to share their data with us because we gave them a discount for driving less. Over time, the data ended up being really valuable to our customers in different ways.
Q: What do individuals need to do to prepare for exposure-based insurance products in the future?
A: Hopefully, products are designed in such a way that there is no new education required for people to be able to consume these products. I actually think it’s the job of the product creator to make something that is seamless and easy to understand, and that the value of it is immediately obvious. It’s important for individuals to have a good sense of what they value from a privacy perspective, and what they’re willing to share or not.
Q: What should organizations do today to make sure that they’re ready to meet customer expectations over the next decade?
A: Many organizations have seen their fundamental infrastructure change so that they can capture a lot of data. But, to this day, a lot of what happens at the organization level is not tracked or captured in any meaningful way. That data can give you a ton of insight into the business and customers you’re trying to build products for.
There is a necessary architecture that you need to think about for the overall business that envisions how you might use data in the future. Because once you don’t have it, and you want to build a product, it’s too late. Organizations will need to have some leadership around what the future state of business would look like in a data-driven company, and then be ready to know what that data is that they need and how to capture it. As long as you’re doing that, you can model it in the future.
Q: What do you think of the idea of an AI as a boss, or more generally, a system managing a company?
A: I’d argue that this is already happening today. Ride-share companies give their drivers maps on where to go and people to pick up—there is no human direction there. If no one showed up at work tomorrow, they would still be running. It’s possible for an entire organization to be managed by an algorithm or a system, but it will be in specific use cases.
Q: What do you envision for Metromile over the next ten years?
A: Over the next ten years, we will continue to extend the pay-per-mile model products into self-driving cars. It will likely be a flexible model and platform that supports manually human-driven miles and autonomous miles. Each mile should be priced differently, and potentially even billed to different places if liability lies with the product manufacturer or the person driving the car. Our model is built to support that, and I expect us to play a big role in that over the next ten years.
Beyond car insurance, exposure-based products are relevant to other lines of insurance. Certainly, there are a number of insurance products that our customers need— homeowners, life, etc.—that could benefit from similar changes to the pricing models. Extending the services that we will offer to other lines of insurance aligns with our mission: to invent smarter ways to manage risk.
This interview is a companion piece to a series of deep dives into the impact that human-machine partnerships will have in four industries over the next decade: health care, financial services, manufacturing, and entertainment and media. For a more detailed explanation on human-machine partnerships, please download the full report, The Next Era of Human-Machine Partnerships, here.
Interview conducted on August 28, 2017 by Institute of the Future.