Ready Yourself for the AI Wave

‍

The AI buzz that crescendoed in 2022 with the release of ChatGPT shows no signs of abating. AI technology continues to present advancing capabilities that impact both personal and professional realms, including audio content generation, video generation, and computer use. These advances are translating to investments, with enterprise AI purchasing up 130% since 2023. In the high-stakes world of insurance distribution, experimentation and piloting with use cases like producer enablement, customer personalization, and more are already underway. Although there are still caution and guardrails desired before widespread implementation, the AI wave is here, even for sensitive enterprise verticals like insurance.

‍

At Chestnut, we have had many discussions with carriers on where to start in their AI implementations. Inevitably, first questions jump to use cases, centering on how the technology can immediately impact producer or customer experience. However, we recommend starting with carrier data foundations.

‍

All AI technologies, including mature machine learning models and hot-off-the-press generative models, are deeply data-driven. As an example, GPT-4 (the secret sauce of ChatGPT) was trained on the equivalent of ninety million novels of textual data from the web to anchor its general intelligence. While the model can impressively ace some standardized exams like the GRE, it cannot reliably execute on insurance distribution needs out of the box. Think of advanced AI models as summa cum laude college graduates. They have immense intellectual potential and diverse subject knowledge, but they still need context to be effective within the insurance industry.

‍

Carriers are in prime position to tune and deepen AI models’ capabilities with their ownership over vast volumes of proprietary data. However, data structure and architecture are major hurdles. To give an example, for an AI to nudge producers towards qualifying for the most relevant bonus programs, the model requires historical performance data from past bonuses for specific regions, enriched by data on producer demographics and characteristics. Cleaning, querying, and cross-referencing all of this data can be a major headache, and this is where Chestnut excels.

‍

As a part of implementation, Chestnut forward deploys engineers to meet your data system where it is. We build a centralized data hub - ingesting and transforming data from your existing architecture to present an integrated source of truth for transaction, policy, and producer data. A reliable and performant data hub unlocks a number of Chestnut capabilities, including intelligent analytics and insights. Just as important, it provides the necessary springboard for any meaningful AI experimentation.

‍

Whether you’re ready to start with a ripple or ride the full wave, Chestnut is here to accelerate your AI efforts.

‍