Take the fight to cyber criminals

Fraud is one of the unfortunate realities that the South African insurance industry must deal with on a daily basis.

Despite insurers detecting more fraud, it is estimated that around £1,9 billion of fraud goes undetected globally each year. The global value of detected fraud in 2011 rose by 7% to £983 million from £919 million in 2010. In 2011, insurers uncovered 138 814 fraudulent insurance claims. This equates to 2 670 claims every week. 

Insurers are desperate to stay ahead of the game when it comes to battling fraud and are always on the lookout for new tactics to use against fraudsters. With the continued growth of technology in the financial services industry, have we finally found the silver bullet that will give insurers the advantage? 

Continued growth

We cannot underestimate the influence that technology will play in the financial services industry in this regard. 

At the recently held South African Fraud Prevention Service (SAFPS) Conference, SAFPS Executive Director Manie van Schalkwyk pointed out that in five years’ time, artificial intelligence (AI) will be a $60 billion industry globally. 

Surely this will offer insurers some kind of advantage when it comes to battling fraud. If we consider that there is a significant movement of criminals towards online platforms, smart technology needs to be where the criminals are in order to identify trends and victims who would be considered soft targets. 

Unfortunate reality

During a panel discussion at the conference, Andries Schutte – a Director at Riskcape – pointed out that technology has created an interconnected world where we constantly live our lives in an interconnected way. 

“The unfortunate reality of this is that criminals are also active in this interconnected world. If we look at South Africa, which is a country that has a significant disparity when it comes to incomes and an economy where there are a lot of people struggling, the landscape is primed for fraudsters. People are turning to fraud because they feel that they have nothing left to lose. This is a very dangerous situation,” said Schutte. 

This is evident when we consider the fact that there is a strong correlation between tough economic climates and increased claims that insurers need to deal with. 

In addition, insurers not only have to deal with fraud, they also need to deal with data theft and identity fraud. Insurers are increasingly becoming data driven organisations and data theft is currently rated as the 4^th largest risk in the world. 

Key combatant

It stands to reason that insurers need to be increasingly vigilant when it comes to addressing the fraud challenge. Unfortunately, insurers cannot be vigilant all the time, and fraudsters are taking advantage of the fact that insurers are not as vigilant as they possibly can be when it comes to the digital battle field.

This is where AI comes to the fore. Insurers can build intelligent models that can monitor the behaviour of fraudsters, identify potential weak spots in an insurer's online system and make important recommendations. This will take the fight to fraudsters. 

“It is important for insurers to use AI to supplement traditional models. AI is only as good as the data that is fed into it. If poor data is fed into the system, then poor outcomes will be the result. There is a huge responsibility when it comes to data and data management. Data scientists are becoming crucial employees within all businesses and we need to cultivate these skills,” said Schutte. 

Social engineering

A key element of engaging over the internet is that a person trusts that they are having an honest engagement. If a client is having a discussion with their broker online, they trust that it is their broker that they are speaking to and not a fraudster. 

Unfortunately, social engineering is becoming rife. Fraudsters are assuming the identity of trusted parties and are garnering vital information from clients which they then use to commit fraud. 

“AI and machine learning can help combat this,” said Ulrich Janse van Rensburg, Head of Retail and Business Bank Fraud Strategy at the ABSA Group. “There is a lot of value in AI and machine learning, but they are not silver bullets. Insurers need to sit down and understand what they want to solve. How are fraudsters impacting their business? What tactics are they using? These are key questions and data regarding this needs to be fed into AI and machine learning models.” 

Key skills

Cyber crime is a highly technical industry and criminals are highly skilled at their jobs. Therefore, insurers need to be just as skilled as the enemy when it comes to combatting cyber crime. 

“There are a lot of people out there who claim that they are experts when it comes to AI and machine learning. But there are skilled people and there are experts. Insurers need to engage with experts in order to develop a system that will effectively combat fraud,” said Janse van Rensburg. 

Editor’s Thoughts:
Technology is the future of the insurance industry; this is not news to anyone. Insurers need to make sure that they are investing in key skills and that they are using refined data to build their AI models. Please comment below, interact with us on Twitter at @fanews_online or email me your thoughts jonathan@fanews.co.za.