Press "Enter" to skip to content

Car insurance and more could get cheaper (and, a lot fairer) thanks to artificial intelligence

TomTom has partnered with AI insurance startup Loop to offer coverage that weighs driving habits over demographics.

Image: iStock/Getty Images Plus

More about artificial intelligence

TomTom unveiled a multiyear partnership Wednesday with Loop, an Austin, Texas-based insurance tech startup, to provide better, more transparent auto insurance prices, the companies said. 

According to the announcement, Loop will provide machine learning data and TomTom, the Netherlands-based mapmaker and location technology specialty firm, will supply Loop with such things as speed profiles, traffic stats and maps services, all in the effort to optimize Loop’s AI capabilities for improving driver safety to help lower insurance rates.

Keeping automobile coverage, and how the rate is calculated, can be frustrating and all too often unfair for consumers. When potential customers inquire about a rate quote for coverage, the most popular legacy insurance companies typically use information such as an individual’s annual income, credit score, educational level and even marital status. And as a result, according to Loop, such metrics hinder low-income drivers with outsized rates and subpar coverage. 

SEE: Hiring Kit: Video Game Programmer (TechRepublic Premium)

Loop, which bills itself as a community-first brand, is hoping to change such circumstances by removing the bias it says is embedded in such standards via a data-driven insurance approach. In particular, it wants to employ AI and telematics — a system where data transmission from the vehicle to a company will keep costs down. How well and where you drive will be key — not a rate that isn’t inflated because the driver doesn’t have a master’s degree or an impeccable FICO score. 

Moreover, TomTom said in a press release that it wants to give Loop AI assistance for analyzing road and driver data, which in turn will help the startup to better understand driver behavior and also encourage those insured with taking safer routes. The payoff would be lowering insurance risk and costs for drivers. 

This data-driven approach for providing consumers a more even playing field is, of course, making its way into an array of business models. For example, according to Vidya Phalke, the chief innovation officer at the risk resilient firm MetricStream, “artificial intelligence is helping companies in financial services when it comes to GRC (governance, risk and compliance) – especially for empowering their front line employees – who are often the first target of cyberattacks.” 

“AI allows companies to break down barriers and silos,” he adds, “by creating a comprehensive view across departments and introducing cognitive search functions so that financial institutions can locate data, and sort and analyze any risk as it develops. When searches are faster, front-line and second-line users can reduce redundancy, and move toward tighter security and smoother compliance.” 

Some experts caution that AI alone cannot remove intrinsic bias. According to AI expert Roman Yampolskiy, a professor of computer science and engineering at the University of Louisville, creating machine learning systems that are 100% impartial is unfeasible. “Researchers usually try to de-bias their AI,” he said, “but there are proven mathematical results showing that some bias is impossible to remove.” 

Still, there is an argument to be made that the use of AI can nonetheless provide financial inclusion for more underserved individuals. For example, Moutusi Sau, Gartner research VP, argues that within the banking services industry, “increasing adoption of AI models in lending would help increase financial inclusion into underbanked populations.” In particular, he maintains in his research that more fairness can come about in the lending process by using AI to help explain the terms of agreements and help banks evaluate a wider population. 

“From a data and model perspective, explainability helps financial institutions identify any systemic bias in model output and retrain as necessary,” he says. While adding: “Although it is not a perfect solution, explainability is a prerequisite to increase diversity and inclusion. To gain the desired outcome, responsibility should be intrinsically included in the algorithm design.” 

And back to the automobile insurance front, Ali Salhi, the chief technology officer at Loop, says that AI-driven data can only benefit drivers — if they are committed to safe driving. 

For Salhi, using TomTom’s map technology and extensive location data will allow his company to “underwrite and rate risk with pinpoint precision that’s never before been seen in private-passenger auto insurance.” As Loop sees it, via algorithms, there is a path to spiking a movement that rattles an industry that brings in $254 billion annually.    

Also see

Source: TechRepublic