Insurance Industry Spends Millions on AI With Minimal Returns as Strategy Gaps Emerge

The global insurance industry is pouring millions into artificial intelligence programs yet extracting almost no value in return, according to new research highlighting a widening gap between technology spending and strategic execution.

A report from Simplifai identified what it called a "fundamental disconnect between AI investment and AI results" across the sector, with the vast majority of carriers failing to move performance metrics despite record-level spending on AI initiatives. The minority of insurers that did achieve measurable gains shared a common pattern: they deployed solutions end-to-end across workflows rather than inserting AI at isolated points in existing processes, and demonstrated meaningful differences in leadership approach and business outcome definition.

The findings arrive as consulting firms and corporate leaders across industries advocate flattening management structures using AI agents. McKinsey partner Alexis Krivkovich and executives at Factory and IBM told Business Insider that AI is enabling leaders to manage wider teams, automate tasks in HR, finance, and legal functions, and accelerate decision-making in what consultants are calling "The Great Flattening."

Meanwhile, major AI firms including OpenAI, Anthropic, and DeepMind are ramping up efforts to build self-improving research systems that automate their own development workflows. Anthropic stated that Claude now authors up to 90 percent of code in some contexts, while OpenAI plans to deploy an AI "intern" within six months, raising concerns about accelerating capability gains and widening regulatory lag.

(The insurance sector's struggles with AI return on investment mirror broader enterprise challenges documented across manufacturing, food service, and other capital-intensive industries that have struggled to scale pilots into production systems.)

The insurance findings stand in sharp contrast to the technology sector's aggressive push toward autonomous systems. Former AI leaders from Microsoft, Google, OpenAI, DeepMind, and the White House warned in interviews published April 2 that AI systems are becoming more capable, autonomous, and harder to control, with risks including deepened inequality, cybercrime, job losses, and concentrated power. They emphasized urgent calls for governance and safety research to mitigate harms.

Simplifai's assessment that "insurance doesn't have an AI problem, it has a strategy execution problem" underscores a pattern emerging across traditional industries: technology acquisition outpaces organizational readiness, and vendors' promises of transformation collide with the messy realities of legacy systems, fragmented data, and misaligned incentives. The carriers that succeeded did not chase increasingly sophisticated technology, the report noted, but instead redesigned leadership structures, workflows, and success metrics before deployment.