Artificial intelligence (AI) is having a profound impact on insurance underwriting, transforming long-established processes and significantly improving insurers’ capabilities. From improving risk assessment to tailoring offers and automating administrative functions, AI is enabling a new era of agility and competitiveness in the InsurTech space, as Earnix explains.
By harnessing AI, insurers can process vast datasets at speed, synthesising complex, often unrelated inputs into a holistic view of risk.
This empowers carriers to create more tailored and competitive policies, priced in line with individual risk profiles and market dynamics. The result is improved decision-making, operational efficiency, and ultimately, profitability.
The technology is already embedded in many underwriting workflows, with numerous property and casualty (P&C) insurers now deploying solutions such as machine learning (ML), large language models (LLMs), and generative AI (GenAI). These tools are supporting a range of use cases across the underwriting spectrum.
One of the most impactful applications of AI is in risk assessment. By analysing structured and unstructured data—from historical claims and customer behaviour to market trends and social media—AI can deliver nuanced insights into potential exposures, enabling more accurate underwriting.
When it comes to policy pricing, AI evaluates individual risk factors to deliver personalised pricing models.
This approach not only makes policies more appealing to customers but also helps insurers avoid adverse selection and improve portfolio performance.
The integration of AI also enables underwriters to automate routine, labour-intensive tasks. AI-powered systems can parse documents, approve applications, and recommend policy terms—all with minimal human input. This reduces costs and shortens the underwriting cycle considerably.
GenAI is emerging as a particularly powerful tool within underwriting. Its applications range from natural language processing (NLP) for document summarisation to generating synthetic datasets for model training—allowing carriers to refine predictive models without compromising customer data privacy.
GenAI also supports chatbot and virtual assistant capabilities, improving both internal productivity and external customer service.
The benefits of AI in insurance underwriting are tangible. Reports show combined ratio improvements of 3–6 percentage points, loss ratio enhancements of 2.1–4.2 percentage points, and portfolio improvements of up to 15%. Additionally, AI-driven pricing personalisation has led to increased gross written premium (GWP) growth of 3–4%.
Productivity gains are equally impressive. McKinsey notes that AI can reduce underwriting costs by up to 30% and increase underwriter productivity by 50%.
In some instances, AI has slashed decision times from several days to just 12.4 minutes, while maintaining near-perfect accuracy.
However, as with all transformative technologies, AI comes with its own set of challenges. Insurers must tackle issues such as data privacy, regulatory compliance, and the potential for bias in automated decisions. Addressing these concerns requires robust oversight, transparent model design, and the integration of human judgement.
Despite these hurdles, AI’s promise in underwriting is unmistakable. It offers insurers a route to more efficient, equitable, and profitable operations.
Carriers seeking to future-proof their businesses should consider investing in AI capabilities now, rethinking their processes and laying the groundwork for scalable, responsible adoption.
Read the full blog from Earnix here.
Copyright © 2025 InsurTech Analyst
