Insurers Spending Billions on AI But Struggling to Show Results
The insurance industry is investing heavily in artificial intelligence without seeing meaningful returns. Research from insurtech firm Simplifai found that 83% of carriers spend at least $5 million annually on AI, yet fewer than 15% report measurable improvements in key performance indicators like combined ratio, cycle time, or loss ratio.
The gap between AI spending and actual results points to a fundamental execution problem. Nearly all insurers-99%-have launched generative AI initiatives, but only 42% have deployed AI in even a single business function. Most applications remain limited to chatbots and document summarization.
Three Barriers Block Scaling
Insurers face three major obstacles: difficulty connecting AI spending to financial outcomes, regulatory complexity, and challenges integrating with legacy systems. These barriers have created what the research describes as "pilot purgatory"-experiments that never progress to full deployment.
The problem is not technical, according to Simplifai's leadership. "Insurance doesn't have an AI problem. It has a strategy execution problem. Technology works. The business model doesn't," the company said.
A Smaller Group Achieves Gains
Some insurers are breaking through by focusing on workflow-driven implementation and end-to-end automation. These firms report productivity gains up to 40% and significant reductions in processing times.
The difference lies in approach. Rather than deploying AI as isolated tools, these insurers embed it into complete workflows across claims, underwriting, or policy administration. This requires governance frameworks and clear operational ownership.
For insurance professionals looking to build AI capability, understanding AI Agents & Automation and AI for Insurance can help bridge the gap between pilot projects and scaled deployment.
Rising Pressure to Scale
The stakes are climbing. Natural catastrophe losses are rising while premium growth is slowing. Insurers that fail to scale AI effectively risk falling behind more agile competitors who deploy it successfully.
The research suggests the next phase of insurance AI will separate winners from laggards-not based on spending, but on execution discipline and willingness to redesign business processes rather than simply automating existing ones.
Your membership also unlocks:
