Magazine

Read the latest edition of AIR and MEIR as an Interactive e-book

Mar 2024

Applied AI in insurance

By -
Source: Asia Insurance Review | Mar 2019

The past few years have been dominated by talk about how new technology trends will change the insurance industry. We look at some examples of how those technologies have been utilised, as insurers and InsurTechs put words to action.
By Ahmad Zaki
 
 
Insurers have long been obsessed with data – analysing it, collecting it and putting it to use in their underwriting and claims processes. Collection of data is possibly the most important step - since the quality of the data will dictate the accuracy of the result. This is why more insurers are stepping into the hardware game, with many offering telematics devices, wearables, and even cameras.
 
One such example is Neos Ventures, a UK-based insurer that uses smart technology to enhance its home insurance offering. It offers alarm systems, indoor security cameras, smoke detection and fire prevention alarms and flood and leak detectors. The company also provides 24/7 monitoring and emergency assistance services. In addition, it offers an iOS mobile application to control the monitoring systems and detect leaks.
 
The idea is that if Neos can provide technology that makes gas leaks, water damage and home intrusions less likely, then they’ll be able to pass along those savings in the form of lower premiums to their customers.
 
Aviva agreed to acquire a majority stake in the company in November of last year, building on Aviva’s existing relationship with Neos. The insurer, through its corporate capital venture fund, Aviva Ventures, announced a strategic investment in Neos in 2017.
 
Pricing based on individual behaviour
IoT sensors allow insurers to move from proxy to source data, which then allows them to price based on the actual behaviour of individuals. This already works for the motor insurance industry, with telematics sensors rewarding ‘safe’ drivers with better premium pricing, and the relationship between wearables and health insurers are well established.
 
Consumers have shown willingness to turn over facial and even biometric data for cheaper products, with one survey by Troubadour Research & Consulting finding that nearly half of consumers would turn over data from wearables to insurance companies. BioBeats and Fitsense are two start-ups tackling wearables data for health insurance, with a focus on personalising employee health plans.
 
The former provides a guided wellbeing course and app that helps customers to understand where their stress comes from and teaches them how to cope with, control and reduce it. With personalised coaching topics and daily goals, BioBeats aims to provide users with a way to track, log and measure their stress and its symptoms. This information is collated in a dashboard, giving users a valuable and easy-to-read insight into their stress, habits and symptoms.
 
Users will be able to find out if their stress is reducing over time, which days of the week they find most stressful, compare locations, activities and interactions which reduce or increase their stress and check their progress as they move through the course.
 
Fitsense works closer with insurers, helping them integrate, process and store data collected from various wearables and apps that might exist across the insurer’s ecosystem. The data is turned into specific customer and risk profiles, and the platform is also able to create personalised white label products, based on the needs of the customer.
 
There is still a fair amount of uncertainty when it comes to usage-based insurance (UBI). A 2017 report from the National Association of Insurance Commissioners noted, “UBI is an emerging area and thus there is still much uncertainty surrounding the selection and interpretation of driving data and how that data should be integrated into existing or new price structures to maintain profitability.” 
 
However, most customers who have used it view it favourably. A 2016 survey by JD Power & Associates found that, “UBI participants provided more positive recommendations and more often indicated that these recommendations resulted in a friend, relative or colleague purchasing from their insurer compared with those customers who did not use a UBI program.” Some insurers offer discounts for participation in usage-based insurance programmes to collect thousands of miles worth of monitored driving data. They can then use this data to benchmark their own risk scoring models on other business lines.
 
However, not everyone is interested in the prospect. The survey found that 21% of customers declined to participate in a UBI programme when it was available and 81% of those respondents did so because they did not want their driving monitored, did not think they would save money, or did not think their premiums would decrease. People with long commutes, who frequently drive long distances or who enjoy speeding on the open road would hardly benefit from their insurance company tracking their behaviour.
 
Buying insurance with a selfie
In January 2017, US-based life insurance start-up Lapetus Solutions made a splash when it offered a service for people to buy life insurance just by using a selfie. Habits such as smoking cigarettes are strong predictors of lifespan, and Lapetus can use facial analysis in their Chronos technology rapidly to assign risk scores without a lengthy or time-consuming medical examination. The company explains their SMILe (smoker indication and lifestyle estimation) approach on their website:
  • Individual surveys collect images, video, demographic and health data;
  • Collections use Institutional Review Board approved protocol with standardised equipment; and 
  • Data collection teams will be engaged for global markets.
 
They also note that multiple points of data on each participant are captured, using a combination of still images and video.
 
Lapetus co-founder and chief data scientist Karl Ricanek said in an interview with USA Today, that there is a unique story for every face. 
 
“Smoking is going to be written on your face,” Mr Ricanek said. “Even if you stopped smoking, once it’s written, it’s there.”
 
It might no longer be possible to lie to an insurer about your smoking habits and get away with it, as the software would be able to detect tell-tale signs on the face, such as crow’s feet around the eyes.
 
The software also estimates body mass index and how quickly a person is ageing biologically, which can be quite different to their chronological age. Mr Ricanek estimates that Chronos would enable a customer to buy life insurance online in as little as 10 minutes without the need for a medical examination. 
 
Lapetus, along with other facial technology developers, claim that advances in life and computing sciences and harnessing the power of cloud computing have allowed for face-reading software to provide more accurate lifespan estimates than traditional methods.
 
Never forget a face
Of course, facial recognition software can be used in much simpler ways as well. Chatbots, which are quickly becoming a staple for insurance customer service, will be able to recognise returning customers and personalise the conversation based on past data. Further, automated personal identity verification can speed authentication necessary for quoting and binding, and machine learning can allow fully online or app-based shopping experience.
 
Image recognition is also at the core of Chinese insurer Zhong An’s business model. The online-only insurance provider has, since 2013, sold 7.2bn insurance products to 429m customers. As it only ever meets its customers online, it has been relying on machine learning to prevent fraud and ensure a personalised customer experience.
 
The end of the status quo
A 2017 survey by Accenture found that the majority of insurance executives believed that “(AI) will significantly transform their industry in the next three years.” The developments over the past year have proven their predictions to be quite accurate and more developments could hit the market before the three year ‘deadline’ is over. A 
 
| Print
CAPTCHA image
Enter the code shown above in the box below.

Note that your comment may be edited or removed in the future, and that your comment may appear alongside the original article on websites other than this one.

 

Recent Comments

There are no comments submitted yet. Do you have an interesting opinion? Then be the first to post a comment.