How property intelligence leverages generative AI & large language models
Oct 2023
Armin MonajemiA deeper dive into CoPilot from Betterview, a virtual underwriting assistant leveraging generative AI and large language models
Oct 2023
Armin Monajemi)
Since this post was originally published, Betterview was acquired by Nearmap. References to Betterview, the company, or its products have been maintained for historical accuracy. Betterview is now the primary solution for the insurance industry at Nearmap.
For two decades, technology has flooded insurers with data, the sheer volume of which has become more overwhelming than it is actionable. Betterview’s mission has been to consolidate data sources in our Property Intelligence Platform to help insurers immediately identify the most relevant insights from mass amounts of information. Recent developments in generative AI gave us an opportunity to further this mission. Using Large Language Models (LLM) through our Betterview Labs initiative, we built Betterview CoPilot, a virtual underwriting assistant.
Betterview CoPilot presents an exciting solution for P&C insurers. Amidst expanding sources of property data, an LLM instantly returns the exact information you need in response to a simple query. With a tool like this, underwriters can accelerate their workflows, freeing them up to spend more time making critical policy decisions.
Betterview is no stranger to artificial intelligence (AI). Through our early adoption of computer vision and machine learning, we developed models that could identify, score, manage, and monitor risk. So, it should come as no surprise our team has responded enthusiastically to recent developments in generative AI/LLMs. Once ChatGPT arrived on the scene, our engineers got to work exploring how the technology could best help our customers.
LLMs turned out to be exactly what we needed to make property data even more accessible to underwriters. Instead of manually sorting through the vast library of property data (including 2,000+ data points sourced from trusted partners,) users can submit a query to our virtual underwriting assistant. Within seconds, Betterview CoPilot returns the necessary information in plain and precise language. Through the power of LLM, underwriters are empowered to ask the right questions, obtain the best possible answer, and take immediate meaningful action. As long as they know what they want to know about a property, Betterview CoPilot will find it for them:
Say, for example, an underwriter is looking at properties in a known wildfire zone. Knowing that Betterview offers a wide range of insights relevant to wildfire risk, they submit a query: “Tell me the wildfire vulnerability score and defensible space for this property.” Or, if they are looking at properties in a coastal, flood-prone area, they might say: “Show me the First Floor Height of this property.” If they are unsure about the age of an addition on a property, they can ask: “Show me the permit history for this property.” Whatever they need to know about a property, Betterview CoPilot is there to tell them.
Thanks to Betterview CoPilot, our expansive ecosystem of property insights is easier than ever to search and deploy. Not only does this conserve time and resources for insurers, but it helps to fulfill one of Betterview’s long-running goals: to make life easier for underwriters. Reach out to us to learn more.
CoPilot is a part of Betterview Labs, our initiative to explore innovative technologies and test their applications for property insurance. Like all Betterview Labs projects, CoPilot reflects our commitment to solving business challenges for insurers.