How big data can combine with AI to insure older vehicles

By Omer Perry

The COVID-19 pandemic may or may not be over, depending on who you ask, but its effects are likely to linger far into the future. And that includes rising car insurance rates.

The pandemic-induced shortage of new vehicles continues to linger, and despite a slight weakening in recent months, demand for used cars remains strong – strong enough to keep used car prices significantly more expensive than they were pre-pandemic. Additionally, as older models remain on the road, spare parts are also in higher demand and become more expensive.

Higher car values and more expensive repair costs mean higher insurance rates – and with inflation pinching consumers, many drivers have reduced their coverage. Especially affected are the high-value comprehensive policies for older vehicles, with customers opting for cheaper non-collision policies.

It's a lose-lose proposition, both for insurers who end up with less business and for customers who end up with less insurance coverage. But advanced technology, including computer vision, big data collected by sensors and artificial intelligence-based analysis, can correct this situation. In-depth analyses of vehicle data, images and driver histories can help companies develop customized policies. This ensures that good drivers who take care of their vehicles over the years are provided with policies they can afford – and will cover the issues they need to have covered – at a reasonable price.

A consumer’s objective in buying insurance is to acquire the highest level of coverage and service possible for the least amount of money. With customers more willing than ever to switch providers – and more complaining that they are overpaying – providers must make their operations as cost-effective and efficient as possible. Advanced technology can help achieve the desired result of better customer service at a better price, the two things that will help retain and even increase customer base.

Presenting policy options and premium costs is the first area that advanced technology can make more efficient. With computer vision and AI, companies can more easily determine the condition of the vehicle they are being asked to insure. Big data analysis is based on a vehicle's history, images of the vehicle (taken by customers themselves using smartphones), data from internal sensors, information on how the vehicle has been and will be used, and images of the vehicle that reveal information that could have even been hidden from its purchaser. With that analysis, companies can make customers a fair offer, providing them the best options for acquiring the best coverage achievable. Setting premiums becomes more efficient and fair. Carriers even can use this technology to provide customized policies tailor-made for specific customers - providing them with the coverage they need and can afford.

After the sale

After the policy is purchased, advanced technologies also can help make claims processing more efficient. In order to pay off a claim, insurers must assess damage, determine how an incident affected the vehicle’s value and ensure that they pay the right amount to the policyholder. To accomplish this more quickly, data gathered from in-vehicle sensors, photos of the damage, and information from traffic cameras, police reports, and more, are uploaded to AI systems. These systems parse through the data and images in order to determine exactly what happened and how much must be paid.

AI-based machine learning systems dig deep into databases to understand the relationship between data and images, providing guidance on the circumstances that led up to an incident. This enables agents to better determine who or what caused an incident, how badly the vehicle is damaged, whether the damage claimed is indeed the result of the incident that is the subject of a claim, and what a fair compensation for the damage should be. As a result, companies ensure that customers get the coverage they need, while ensuring that they do not fall victim to insurance fraud.

Although this system will help companies issuing policies for new vehicles, it's perhaps even more essential for policies issued for used vehicles – where there is a much greater likelihood of damage or other problems in a vehicle's history. Human agents (subject to human error) conducting a visual inspection of a vehicle could easily miss details, with the company either assessing a vehicle's condition incorrectly and charging too much or too little for coverage, or underpaying/overpaying a claim. The system also can help companies assist customers who have been in an accident, providing them with the best coverage options available for the money they are willing to spend.

AI-based data analysis, along with advanced images produced by computer vision systems, can thus help insurers build a better relationship with customers and provide more accurate results for everyone involved. Advanced technologies can help build a trusting relationship with customers by helping them offer coverage and set premiums that customers can afford or evaluate payments after incidents.

With advanced technology, insurers can protect their interests while making their customers happier – a win-win for both sides.

Omer Perry is product manager at Ravin AI. She may be contacted at [email protected].