close up man hand press on display screen tablet turn on wifi technology for using home automation … [+] function to looking inside or outside home and property , ai innovation future concept


How are big data and AI changing the insurance industry? originally appeared on Quora: the place to gain and share knowledge, empowering people to learn from others and better understand the world. 

Answer by Peter Colis, CEO and Co-founder at Ethos Life, on Quora: 

Big data is shaping the next wave of insurtech technology and has already helped us build better products and experiences. However, the insurance industry as a whole is tasked with striking a careful balance between the desire to keep innovating with the need to remain vigilant about how and when consumer data is used. As available data sources multiply, there’s a growing conversation around the types of online data providers use to assess an individual’s risk. This becomes even more thorny amidst the backdrop of increasingly strict data sharing and privacy regulations.

It’s increasingly common for life insurance companies to use “non-traditional” sources of data — such as credit scores, court documents, and motor vehicle records — in assessing risk. To keep this practice ethical and avoid becoming invasive, it’s on insurance industry to find ways to innovate while still putting public good first.

Beyond the public data sources I’ve already listed, fitness trackers and wearables are an emerging source of risk data in the insurance world. While sharing your fitness tracker data could benefit some people applying for life insurance policies (as a way to prove physical health), it also poses the risk of creating inaccurate or biased data pools. The nature of an opt-in program could negatively impact those who choose not to participate, incentivizing giving up one’s right to keep that data private. Additionally, the data from those who do opt-in may not provide an honest representation of health status. Studies have shown many fitness trackers have error rates of 10 to 20 percent.

Finally, social media data is another contentious topic. Using information derived from people’s social media profiles, like publicly accessible Instagram accounts, is a slippery slope when it comes to assessing risk. It’s not a consistent source of client data and it can chip away at consumer trust. Ultimately, the data you’d glean is likely not even substantive or actionable enough for quality underwriting.

Insurance will continue to evolve into an increasingly data-driven industry. Already, it’s how we make informed decisions on anything, from marketing to underwriting. But at the end of the day, we can never lose sight of the main objective: protecting families.

This question originally appeared on Quora - the place to gain and share knowledge, empowering people to learn from others and better understand the world. You can follow Quora on Twitter and Facebook. More questions: