How to utilise Artificial Intelligence in fraud claims

AI can be used to combat fraud, but it also requires a certain mindset

How to utilise Artificial Intelligence in fraud claims


By Alexi Demetriadi

Nigel Cade, managing director of The Insurance Claims Service Centre, explained to Insurance Business the innovative ways in which Artificial Intelligence (AI) can be used to help detect and prevent fraudulent claims, but also that a ‘fraud fighting culture’ needs to be present to properly utilise the software.

“There is often a zero-tolerance policy, but not necessarily a fraud fighting culture in place,” Cade explained. “I don’t think that the industry is yet to deal that well with fraud at all.”

Cade spoke to Insurance Business in anticipation of his presentation on the topic at the TechFest in May. Fraud, he said, was still a pressing issue that the insurance industry must grapple with.

“It’s huge – it’s a huge problem still,” he said. “Sixty per cent (60%) to 70% of claims that we deal with have an element of fraud in them.” And as new ways of living our lives pop up, so too will fraud. Data mining, identity theft and fraud associated with social media will become even more widespread in the coming years, Cade said.

One piece of technology which could have a significant impact in the ongoing fight against fraud is AI.

“AI is brilliant,” said Cade. “It makes processes better, quicker, fast and clearer.”

So, how can AI help insurers? Insurance Business has reported in the past on how insurers are using AI to boost their business in different ways; from accelerating processing times to producing better, tailormade policies for clients.

But how can it help combat fraudulent activity in the claims space? Careful not to completely divulge his whole presentation, Cade says one way is to look at the work AI can do in the imaging space. “Analysation of images is one of the strongest examples,” he said.

He recalled the story of a ring, a bag and the successful usage of metadata imagery.

“I used to do a lot of work on meta and digital data,” Cade recalled. “And we received one claim that involved an apparent lost handbag and ring.” Cade asked the insurer to send over some photos of the items to him and his team. “Using the digital footprint, we were able to get into the metadata of the photos,” he explained. “And what we found was that the photos were actually taken about a month or so after the loss claim was filed – the ring was still on her finger and the bag still on her arm.”

Cade is adamant, however, that no matter how useful AI can be in helping deal with fraud, it is still only a piece of software that needs to be used by an engaged and aware operator.

“AI is just a tool – to make it useful, insurers need to create a policy and culture at senior management level that is taken up by the whole company – a fraud fighting culture,” said Cade. “A culture like that requires communication between all departments as well as fraud awareness training and a buy-in from senior staff – and then when you get that culture, you can leverage AI.”

To learn more about fraud in the insurnace sphere, and to hear from other industry figures on tech-related topics, sign up to Insurance Business’s TechFest in Sydney on May 14.

Related Stories

Keep up with the latest news and events

Join our mailing list, it’s free!