Imagine a world where AI chatbots revolutionize the insurance industry, promising to save agents precious time. But is it all it's cracked up to be?
Researchers from Dakota State University, in collaboration with Safety Insurance, developed an experimental chatbot named Axlerod to assist independent insurance agents. While the potential benefits are intriguing, the actual impact might not be as groundbreaking as one would hope.
According to a pre-print paper, Axlerod can help auto insurance agents with specific information retrieval tasks, resulting in an average time savings of 2.42 seconds per search. This includes tasks like finding client policy numbers, determining AutoPay eligibility, and identifying covered vehicles.
But here's where it gets controversial... The value of this assistance varies greatly depending on usage and the specific needs of insurance agents. It underscores the challenges businesses face when implementing AI systems.
DSU researchers argue that business-facing chatbots are more practical than consumer-facing ones, given insurance agents' expertise in interpreting policy language and identifying potential AI hallucinations. They believe AI technology can bring about significant changes in the industry.
"This study showcases AI's transformative potential, demonstrating how Axlerod-like chatbots can enhance operational efficiency for insurance agents," state the authors.
However, the term "transformative" might be a stretch when it comes to the actual time saved. On average, agents save a mere 2.42 seconds per search, which, when repeated 20 times a day, amounts to a total daily savings of 48.4 seconds, or 12,100 seconds over 250 business days.
And this is the part most people miss... When considering the cost of implementing AI systems, such as Google Gemini Business edition, which starts at $21 per month, the return on investment becomes a topic of debate. With an average insurance agent salary of $80,000, the daily savings of 48.4 seconds translates to a meager $121 in labor costs over a year.
Khandaker Mamun Ahmed, an assistant professor at Dakota State University and co-author of the paper, believes AI has strong potential to transform insurance operations. He notes that with an AI system directly accessing relevant data sources, the number of information lookups could be significantly higher, reducing retrieval friction.
But is it worth it? When considering the cost per search ($0.0075) and the potential daily savings ($1.936), the daily ROI can be substantial (222%). However, the break-even point is reached at just 0.31 searches, indicating that the system might not be cost-effective for low-usage scenarios.
Scott Johnson, an independent insurance broker, shares his perspective, stating that while he makes policy queries about basic data 10-20 times a day, he's cautious about AI chatbots. He believes existing software like EZLynx already automates many tasks, avoiding 80% of inquiries.
Ahmed acknowledges the need for robust field testing and expanded evaluation with practicing agents to validate real-world utility and assess agent trust.
So, while AI chatbots like Axlerod show promise, the actual impact on insurance agents' daily routines and the industry as a whole remains a topic of discussion.
What are your thoughts? Do you think AI chatbots will revolutionize the insurance industry, or is it just a lot of hype? We'd love to hear your opinions in the comments!
