Insurers Chase AI's Promise, But Can They Earn Our Trust?

BOSTON, MA – June 03, 2026 – The insurance industry, a sector built on centuries of risk assessment and human judgment, stands at a profound inflection point. The buzz around artificial intelligence has moved from the experimental fringes to the operational core, promising unprecedented efficiency in underwriting, claims, and customer service. But as insurers rush to adopt these powerful tools, they face a more fundamental challenge: earning the trust of regulators, customers, and their own boards.

This high-stakes balancing act is the central theme at this week’s Insurtech Insights USA conference, where industry leaders are grappling with how to scale AI without sacrificing governance. Highlighting this critical conversation is a mainstage session featuring Duck Creek CEO Hardeep Gulati, titled, "No Trust, No Scale: The Executive Playbook for Trusted AI Decisioning in P&C Insurance." The title itself is a stark admission of the industry's core dilemma. The promise of an $80 billion annual impact in the U.S. market alone is a powerful lure, but the path to that prize is paved with ethical and regulatory minefields.

Duck Creek, a major provider of core systems for the property and casualty (P&C) market, is positioning itself as a guide through this complex terrain. The Boston-based firm is using the conference to showcase its new Agentic AI Platform, a suite of tools designed to embed intelligent automation directly into the insurance lifecycle. But as the industry moves past disconnected copilots and simple automation, the question becomes less about what AI can do and more about how it does it—and who remains in control.

The 'Agentic' Evolution: Beyond Simple Automation

For years, AI in insurance was synonymous with automating discrete tasks. Now, the industry is embracing a more sophisticated concept: "agentic AI." This refers to systems where AI agents are not just assistants but active participants, capable of orchestrating complex, end-to-end workflows. Duck Creek's new platform embodies this shift, aiming to transform core processes like underwriting and claims.

Two of its flagship applications, the Agentic Underwriting Workbench and Agentic First Notice of Loss (FNOL), illustrate this evolution. The underwriting tool uses AI agents to intake, triage, and enrich submissions in real-time, automating data gathering to present underwriters with decision-ready files. The goal is to slash quote times and improve risk selection. Similarly, the FNOL application, developed with Google Cloud and its Gemini models, coordinates agents across voice, mobile, and digital channels to intelligently capture and validate claims information from the very first contact, even flagging potential fraud.

What sets these systems apart, according to the company, is a hybrid approach called neuro-symbolic reasoning. This method pairs the probabilistic power of large language models with the deterministic, rules-based logic that governs the highly regulated insurance industry. In practice, this means AI agents can reason and act, but only within the guardrails of an insurer’s specific compliance and business rules. It’s a move toward granting AI more autonomy while ensuring it remains tethered to human-defined controls.

This focus on practical, grounded application is resonating with carriers. "Modern insurers need AI solutions that are not only powerful, but trusted and grounded in real operational workflows," said Ramana Narayanam, Head of IT at Coaction Global, who will join Gulati on stage. His perspective reflects a growing consensus: for AI to be successful, it must be seamlessly woven into the fabric of daily operations, strengthening data foundations and supporting both business agility and governance.

A Crowded Field Chasing the Same Ideal

Duck Creek is not alone in its pursuit of trusted, agentic AI. The entire core systems market is converging on this strategy, turning it into the next major competitive battleground. Major rivals like Guidewire, Majesco, and Sapiens have all announced significant investments and product launches centered on embedding governed AI directly into their platforms.

Guidewire, a market leader for large P&C insurers, recently launched Guidewire ProNavigator, an AI assistant embedded across its core suite to provide role-specific insights for underwriters and claims adjusters. Majesco is quadrupling its AI development budget in 2026, rolling out its own Majesco Copilot and a suite of AI agents. Sapiens is accelerating its "Insurance Agentification" program with dedicated agentic products for claims, underwriting, and policy management, even establishing AI Customer Experience Labs to test applications.

This industry-wide pivot reveals a crucial insight: the era of AI as a bolt-on feature is over. The future belongs to AI-native core systems where intelligence is an inherent part of the architecture. The common thread in all these announcements is the emphasis on governance, security, and auditability. The race is not just to build the smartest AI, but the most responsible one. As Gulati noted, "Every insurer wants the upside of AI, including faster underwriting, smarter claims and better operations. The difference is whether they can deploy AI with the governance their regulators, customers and boards demand."

Navigating the Regulatory Gauntlet

That demand for governance is no longer abstract. Regulators are moving swiftly to establish clear rules for AI use in insurance, transforming ethical principles into legal requirements. In the United States, the National Association of Insurance Commissioners (NAIC) has been at the forefront. In December 2023, it adopted a Model Bulletin on the Use of Artificial Intelligence Systems by Insurers, providing guidance on how to ensure AI complies with existing laws against unfair trade practices and discrimination. As of late 2025, 24 jurisdictions had adopted the bulletin, and the NAIC has since launched a pilot program to test an AI evaluation tool with insurers in nine states.

This state-level action is complemented by global trends, most notably the European Union's landmark AI Act, which classifies AI systems by risk and imposes strict obligations on those deemed "high-risk"—a category that will almost certainly include underwriting and claims decisioning systems. These frameworks are coalescing around a core set of principles: fairness, transparency, and accountability.

Insurers are now expected to develop comprehensive AI governance programs that detail how their systems work, mitigate consumer risks, and maintain clear audit trails. The concept of a "human-in-the-loop" is becoming a mandate, ensuring that for all the speed and power of AI, a human expert remains able to validate decisions, identify bias, and provide ultimate accountability. This regulatory pressure forces technology providers to build platforms with "AI Assurance" at their core, offering the traceability and compliance controls that turn a black box into a transparent, auditable system. For an industry whose currency is public confidence, proving that its technology is not only intelligent but also fair and accountable is the only path to sustainable innovation.