Is AI shaping the future of insurance fraud investigations?
By now, most of us are familiar with Artificial Intelligence (AI), the software that powers many of our latest gadgets and devices day-to-day.
From the digital personal assistants we rely on like Amazon Echo to the experiences in video games, AI is a integral part of everyday life to many of us.
However, could it be better used to solve a more serious problem? In our world, insurance fraud is a big issue.
There are some existing tools that can be used to combat it. One is the Insurance Fraud Register, which holds details of individuals who have been associated with fraudulent activity in the past. However, it only has data going back to 2013, so there is room for error.
What’s more, just 62% of general insurers use the system, meaning a large number of criminals can potentially slip through the net. It’s an expensive problem: research from the Association of British Insurers (ABI) shows that in 2016 there were around 125,000 fraudulent claims amounting to £1.3 billion.
The research also suggests that a similar amount of frauds go undetected every year. Car insurance scams like “crash for cash” frauds are commonplace. Detecting just 1% of these claims could save the industry £13 million.
But an AI-led solution could be just around the corner.
AI company WeSee is developing a system that can analyse involuntary vocal and facial responses including pupil dilation, tone of voice and eye movement, to detect dishonest or suspicious behaviours.
David Fulton, CEO of WeSee, said: “Using advanced deep learning techniques, the system would analyse an individual’s responses and micro-emotion reactions to a set of questions in real time and deliver an assessment of their veracity to the insurer almost instantly."
Detecting just 1% of these claims could save the industry £13 million.
He says this will help insurance claims handlers “do their job more effectively by flagging up clearly the most suspicious claimants for further, more detailed investigation.”
Technological developments like this can benefit the insurance industry and our customers tremendously, potentially reducing costs by saving time and money for everyone and catching more of those who try to commit insurance fraud.