Feb 28 · 5 min read

The three AI-driven improvements we’ll see in the insurance industry by 2020

Artificial Intelligence (AI) technologies are already transforming our lives with connected devices (IoT) and the proliferation of data associated with them. Deep learning, IoT and machine learning are becoming ubiquitous, allowing insurers to collect and analyze more data faster and more accurately than ever before. So it’s little wonder McKinsey has predicted that the insurance industry is on the verge of a “seismic, tech-driven shift” to AI. That shift will have significant implications for nearly every part of the insurance industry’s operations from product development to claims processing. So how will all that turbo-charged data-crunching manifest? We have highlighted three key areas where the insurance industry will see a significant advancement thanks to AI-driven technology’s ability to transform actuarial science.

  1. Faster, streamlined processes

Onboarding new customers and managing claims consume considerable resources in regards to time, personnel and risk for an insurer. AI’s ability to collect and analyze unprecedented amounts of customer data will lead to significant workflow improvements saving money and resulting in better, faster customer experience. Blockchain enabled smart-contracts, and digital identity services can provide immutable, reliable customer information, while robot agents can now manage the more transactional side of customer service interactions using natural language processing. At claim time, parametric insurance modeling is simplifying the loss investigation process which can give customers confidence when it comes to liquidity and speed of pay-out.

A recent Accenture study found US insurers could save up to $7 billion in 18 months using AI-driven technologies. When it comes to revenue growth, AI technology can identify preference patterns based on customer interactions to cross-sell and upsell products or services. With benefits like these, it’s no surprise that we’re seeing all the major insurers experiment with AI-led processes. Liberty Mutual is reportedly developing an app to help drivers to assess car accident damage in real-time with their smartphone camera. Computer vision can be used to determine the severity of damage using real-time video footage. The app will use AI trained on thousands of images from car crashes to provide damage-specific repair estimates quickly.

Not only will AI expedite the speed and accuracy of claims, but it will be able to help prevent accidents too. Insurance start-up Lemonade, claims its AI-driven app can insure someone in 90 seconds and pay out in three minutes. Last month, the online insurer announced more than $57 million in sales for 2018 and plans to expand into Europe.

2. Eliminating insurance fraud

Insurance fraud is estimated to cost the US insurance industry more than US$40 billion per year. But it’s not just insurers picking up the tab, the cost of fraud is passed onto customers in their premiums. AI-driven technology, however, can catch clients who are a little too liberal with the truth at claim time. It is currently possible to combine AI and voice recognition technology to detect and interpret emotion and linguistics to assess the credibility of insurance claims.

AI can also reward good behavior resulting in a fairer allocation of premium costs. Last September, US life insurer John Hancock announced its interactive Vitality program, which involves optional fitness tracking through wearable devices and smartphones, would be part of all policies. The basic Vitality program requires customers to log their activity in an app or website and on reaching fitness milestones, customers receive gift cards and other incentives. The expanded option provides customers with wearable devices and discounts on premiums along with other benefits.

With cars becoming more digital, and our phones monitoring our every move, the data insurance companies can ask customers for is immense. While some may find sharing their every step with their insurer intrusive, the lure of cheaper premiums will appeal to many.

3. Tailor-made products and customer service

With an increased ability to calculate everything from the safest route to work to an individual’s life expectancy on any given day, insurers will be able to tailor products and premiums not just to a specific demographic, but to the individual. AI can provide solutions that understand natural language, learn and build knowledge, and communicate seamlessly with customers. Advances in emotion recognition and sensing technology will allow insurers to analyze patterns in customer behavior allowing them to identify and respond quickly to any new customer requirements.

With one survey finding approximately 41 percent of respondents left an insurer due to a poor customer experience, insurers are increasingly utilizing third party tech providers to gain specific customer information quickly and efficiently. European telematics provider Octo Telematics has analyzed 228 billion miles of driving data and correlates a policyholder’s behavior and driving data against similar driving profiles in their data-set to form a risk profile. The company claims it makes insurance pricing fairer, rewarding good behavior while providing insurers with an accurate risk and predictive profile. That sort of customization used to come at a price, but not anymore. Cape Analytics’ AI technology uses satellite images of property to provide packages and pricing tailored to an individual’s property without the cost of sending an assessor.

Given that actuarial science is a data-driven field, and some say that actuaries were the original data scientists, it’s not surprising that AI-driven analytics will transform the insurance industry. Considering the data required for AI-driven solutions available right now as well as the significant benefits to both customers and insurers, how will existing insurers evolve to meet customer demand and compete with AI-driven startups?

While offering significant advantages, AI requires enormous amounts of compute capacity, which is both expensive and time-consuming to develop. Rather than attempt to create the sort of capacity needed to train and maintain AI in-house, organizations are utilizing specialist compute providers such as the decentralized compute platform we’ve built at Tatau. Tatau has harnessed a distributed network of GPUs to create a giant supercomputer. The Tatau platform taps into the latent existing compute rather than building more expensive, resource-heavy hardware, to provide organizations worldwide with a cost-effective, environmentally-friendly and — most importantly — scalable source of compute.

One thing is for sure: AI-driven technology is already transforming the insurance industry. Like any other significant development, developing timely evolution, the right skills, and scalable capacity will ensure both established institutions and start-ups alike reap the business and customer-service benefits that AI-driven technology can deliver.


Distributed supercomputing platform


Written by

Tatau is a blockchain-based, distributed platform that allows anyone to buy and sell computational capacity on decentralized, GPU-based machines



Distributed supercomputing platform

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade