Verisk integrates GenAI into its PAAS platform, streamlining audits by automating data retrieval, and reducing manual workload.
Verisk has transformed premium audits with PAAS AI, a GenAI-powered assistant designed to provide real-time, accurate responses to complex insurance queries. By leveraging retrieval-augmented generation (RAG) and Amazon Bedrock’s large language models, the AI significantly reduces the time auditors spend searching for information. This innovation enables insurers to classify risks faster and with greater precision. Minimizing errors in workers’ compensation, general liability, and commercial auto policies.
One of the biggest challenges in premium audits is navigating vast amounts of regulatory documents and classification guides. Traditional searches are slow and often yield incomplete or inconsistent results. PAAS AI eliminates these inefficiencies by dynamically retrieving the most relevant information from over 40,000 classification guides and 500+ bulletins. The AI assistant refines search accuracy while minimizing hallucinations. Ensuring responses are grounded in verified data.
The AI’s architecture includes Amazon OpenSearch for rapid document indexing and Snowflake for scalable data analytics. Its integration with conversational AI enhances user interactions. Enabling auditors to receive precise, context-aware recommendations instantly. The system continuously improves through a built-in evaluation API. Which scores responses based on relevance, accuracy, and faithfulness to the retrieved data.
Verisk’s PAAS AI has already reduced processing times by up to 98%, allowing auditors to shift their focus to high-value tasks. As the company expands these AI-driven capabilities, future iterations will proactively suggest optimizations and integrate deeper automation into the audit workflow, further redefining efficiency in the insurance industry.
