Recently I attended a round table event hosted by Accenture that looked at the changes occurring in the insurance industry. The discussion was interesting, not least for the apparent blindness in the industry to non-traditional competitors who increasingly have access to huge databases surrounding our behavior, and the AI based intelligence to make sense of that.
In an industry that is largely founded on an understanding of risk, these factors seem considerable. A glimpse into this potential future comes via a Japanese personal insurance company who are replacing 34 employees with an AI system based upon IBM’s Watson.
The system will support the company by reading medical certificates written by doctors to collect the kind of information required to process claims, including medical histories, length of stays and the treatment they received.
Early days
Suffice to say, this project is a very early-stage example of what might be possible, and much of what Watson is doing is work that has historically been done by computers. The data is processed by the machine, but the human still makes the final call.
The possibilities are considerable however, especially as non-insurance companies gain access to data around your behavior. For instance, if Nest provides Google with data around your home, what odds that they begin to offer home insurance? What odds that car companies use the data your vehicle records around your driving habits to offer personalized insurance to you?
As machine learning becomes more capable, the costs involved in offering this kind of personalized service will drop considerably, thus rendering those with access to the data the kings of the castle.
“AI’s initial impact primarily relates to improving efficiencies and automating existing customer-facing, underwriting and claims processes,” Price Waterhouse Cooper said in a report about AI in insurance. “Over time, its impact will be more profound; it will identify, assess, and underwrite emerging risks and identify new revenue sources.”
Ownership of data
With so much competitive advantage derived from data, the next battle, one would think, might be over just who owns that data. The last few years have seen technologies from Facebook to Gmail offered freely to users on the grounds that the data then belongs to the providers.
When that data starts to become more personal however, the discussion moves to another level. For instance, how might patients respond to their health data being arbitraged by insurance companies? This is especially so when that data is increasingly user generated, whether it’s through mobile apps, wearable devices or even personal genome sequencing.
Much of the initial forays of the industry into AI have been around fraud prevention via pattern matching in huge chunks of data. The potential is evident as the amount of data around our activities grows exponentially, but whether we will cede control of that data, and on what terms, remains to be seen.
