AI in Property Claims: Progress Is Real, but Scalability Is Not

"Only 7% of property insurance carriers have achieved scalable AI success in claims operations." Source: Risk & Insurance
The promise of artificial intelligence in property insurance claims is undeniable. Faster processing, improved accuracy, and reduced fraud detection times are just a few of the benefits that have driven a surge in AI experimentation. Yet, despite this enthusiasm, the path to real-world scalability remains elusive. A recent Sedgwick report reveals a sobering truth: only 7% of property insurance carriers have achieved scalable AI success in claims operations. That means 93% are either in the testing phase, using AI in isolated pockets, or relying entirely on traditional methods. What explains this slow progress? Three key factors stand out. **1. Fragmented Use Cases** Many insurers are using AI for specific tasks—like photo analysis or document classification—without integrating it into broader claims workflows. These siloed applications offer incremental improvements but fail to deliver the systemic efficiency gains needed for true transformation. In fact, 42% of carriers still use AI in less than 20% of claims, according to the report. **2. Infrastructure Gaps** A lack of digital infrastructure is another roadblock. Legacy systems, inconsistent data quality, and poor interoperability between departments hinder AI deployment. For example, 63% of insurers report that poor data governance is a major obstacle to AI scaling. In property claims, where data sources range from satellite imagery to adjuster notes, clean, structured data is essential. **3. Talent and Change Management** Even when the technology is in place, organizational inertia can slow progress. Only 28% of carriers have dedicated AI teams, and many employees remain skeptical of automation. This cultural resistance is especially pronounced in claims, where human judgment and customer empathy are still seen as irreplaceable. To close the gap between innovation and impact, insurers must move beyond point solutions and invest in holistic AI strategies. A useful comparison shows how far the industry still has to go: - Early Adopters (7%): Full AI integration across 80%+ of claims, with measurable cost and speed reductions. - Pilot Stage (22%): AI used in 10–40% of claims, often in controlled trials. - Exploratory Stage (45%): AI used in <10% of claims, mostly for data collection or proof-of-concept. - Non-Users (26%): No AI use in claims operations, often due to resource or strategic constraints. The path to AI maturity is clear but challenging. Insurers must clean and unify their data, build cross-functional teams, and rethink legacy processes. Until then, the 7% figure will remain a benchmark—and a challenge—for the rest of the industry. Can AI revolutionize property claims? Yes. But only when it’s not just a tool, but a transformation.