Grant Thornton: Insurers See AI Gains but Face Governance Gap

Insurers are seeing gains from artificial intelligence, but many are also exposing themselves to unnecessary risks, according to a new Grant Thornton report.

The independent CPA firm’s 2026 AI Impact Survey Report found that 44% of insurance executives say governance or compliance challenges have contributed to AI project failure or underperformance. Only 24% are very confident they could pass an independent AI governance review in 90 days.

“Without clear policies and tested controls, insurers are leaving their organizations open to risk with regulators and customers, fueling financial pressure that could ultimately erode product profitability,” Grant Thornton said. “Tested and provable governance gives insurers confidence to deploy AI across higher-value workflows, leading to improved ROI and revenue growth.”

Many of the 100 insurance respondents to the survey reported positive effects of AI. For example, 52% of insurance execs said AI has driven revenue growth, 62% reported improved decision-making insights and half said it has reduced costs.

While most insurers have AI governance policies in place, the report found many lack the operational infrastructure to prove and test them. In total, 61% of insurance executives said their boards have established AI governance policies, but even when controls exist, evidence is fragmented across teams and tools—an issue that often stems from poor guardrails and decision rights.

“Insurers that are successfully managing their AI-related risk define, assess and classify AI use cases and prioritize risk based on their potential impact and complexity,” the report said.

The findings echo other industry research.

In a separate report released April 27, AM Best reported that while nearly 60% of respondents to a survey it conducted expect AI to significantly transform their business models within one to three years, obstacles to deploying AI within their organizations remain. The largest impediments are data readiness, security and privacy and integration with legacy systems.

“AI systems can produce unreliable outputs when underlying data is of poor quality, fragmented across legacy systems, insufficiently governed or lacking appropriate context,” said Sridhar Manyem, senior director, industry research and analytics at AM Best. “Insurers that have invested in modernizing their legacy systems and have robust data governance will find it easier to integrate AI into their workflow.”

According to responses from more than 150 rated insurers and MGAs with a Best’s Performance Assessment, insurers are rapidly deploying AI. Forty-one percent said their organizations are actively using AI across core business areas. Nearly 20% agreed or strongly agreed that their organization is at an advanced stage of implementation.

Grant Thornton encouraged insurers to evaluate current governance structures and modernize as needed, assess their operating models for greater AI adoption and build customer and regulator trust—not just revenue growth.

“In today’s competitive insurance landscape, insurers need to provide measurable value of their AI investments—quickly,” the firm said. “But governance is what makes AI revenue scalable, defensible and sustainable.”