FOLLOW US

(USA Herald) - The insurance industry is experiencing a digital revolution, with artificial intelligence (AI) playing an increasingly role in underwriting, claims adjusting, and other key functions. However, this hasty embrace of AI is raising concerns about fairness, transparency, and bad-faith in the claims handling process.

As Samuel Lopez, an investigative and legal analyst for USA Herald, explains, "Policyholders assert that using software expressly designed to favor the insurance company is irreconcilable with the duty of utmost good faith."

Highlighting these concerns, a webinar hosted by Strafford Publications on March 21, 2024, featured panelists who examined the grounds for bad faith claims and the substantive and evidentiary challenges stemming from the widespread adoption of AI in the insurance industry, indicating that significant concerns remain. Bad Faith and Generative Ai in the Insurance Sector

The Rise of Insurtech and AI-powered Claims

Insurtech, which stands for “insurance technology,” is a trend in the insurance industry that uses technology to allegedly enhance efficiency and customer experience. It’s known for using data analysis, the Internet of Things (IoT), and artificial intelligence (AI) to create more competitive pricing models and to process claims and policies more effectively. However, like any technology-driven sector, it faces challenges such as regulatory issues, and the potential for bias in its output.

AI and Potential Bad Faith Claims

There are numerous lawsuits developing that allege insurers are using AI in ways that violate their duty of good faith towards policyholders, and third-party claimants. These claims focus on several key concerns:

• Bias and Discrimination: AI algorithms can perpetuate historical biases present in claims data, leading to unfair denials or low-ball settlements for certain people or demographics. An example of AI algorithms perpetuating historical biases can be seen in the case of Amazon’s recruitment algorithm. In 2018, Amazon had to stop using a recruitment tool that was biased against women. The algorithm was trained on historical hiring data, which disproportionately included men, leading it to favor male candidates over female candidates. This is a clear instance where reliance on historical data, without correcting for existing biases, resulted in unfair treatment of a specific demographic group. (Bias and Discrimination at Amazon)
• Ai Used in Claims Processing Similarly, AI used in claims processing has the potential to be trained on historical claims data that reflects past prejudices or disparities. If certain demographics were historically under-compensated or claims from these groups were denied more frequently, the AI might learn to imitate these patterns, leading to ongoing unfair treatment.
• Lack of Transparency: The inner workings of many AI systems are shrouded in secrecy, making it difficult for policyholders to understand how claim decisions are made. An example of the lack of transparency in AI systems can be seen in the case of ride-share companies like Uber. In 2020, former drivers brought an action against Uber, alleging that they had been accused of fraud and had their contracts terminated based on algorithmic decisions that were not sufficiently transparent. The drivers claimed they could not understand the decision-making process, which led to claims of discrimination and other harmful effects. This case highlights the challenges policyholders and Third-Party Claimants may face when AI systems make decisions without clear explanations, making it difficult for those affected to understand or contest those decisions. (Uber Lawsuit).

Contractual and Extra-contractual Claims