Across the global insurance landscape, leaders are confronting a pivotal challenge: turning AI from a series of experiments into a scalable force that delivers enterprise-wide impact.

Momentum is clearly building. AI investment is accelerating across industries and insurers are among the most committed. According to CGI's Voice of Our Clients research, 24% of executives interviewed have ecosystem-wide, holistic AI strategies. At the same time, many organizations are already progressing beyond pilots, embedding machine learning into underwriting, automating claims workflows and introducing generative AI into day-to-day operations.

Yet increased adoption does not automatically translate into transformation.

While 46% of insurance executives report achieving expected results from their digital strategies, only 43% have an enterprise-wide AI strategy. This gap highlights a critical challenge: scaling AI requires more than deploying tools. It demands structural alignment across the organization. The next phase of AI maturity will be defined not by additional pilots, but by the strategic decisions insurers make to scale responsibly and deliver measurable outcomes.

Five key insights define the current landscape:

1. Beyond process automation and assistance: redefining the operating model

Expectations for AI-driven value are high, but realizing that value requires more than automation alone. In insurance, outcomes are driven by interconnected workflows, not isolated tasks. The impact of AI will ultimately be measured by its ability to:

  • improve underwriting quality and risk selection
  • reduce friction in claims
  • strengthen fraud detection
  • enhance transparency
  • deliver better customer experiences. 

Achieving these outcomes consistently at scale requires transforming end-to-end workflows across the operating model. Today, most insurers remain in assistive and decision-support stages of AI maturity. Tools summarize files, recommend next-best actions and accelerate individual tasks. These improvements matter, but they do not redefine the customer journey.

True transformation occurs when AI orchestrates processes across the full value chain, linking underwriting, claims and pricing into closed feedback loops. This requires rethinking how work is performed, not simply speeding up existing steps.

Without measurable customer and operational outcomes, AI risks being layered onto legacy systems and processes rather than reshaping them. As insurers begin to rethink their operating models, new considerations emerge around data, trust and control.

2. Sovereignty is a strategic differentiator

As insurers scale AI across core workflows, trust becomes a defining constraint. Claims environments process highly sensitive personal information, including health records, financial data and accident details, within a complex and evolving privacy and regulatory landscape.

CGI’s Voice of Our Clients research shows that 70% of executives rate political and regulatory shifts as high-impact, and that cybersecurity remains the top initiative for mitigating industry risk. Traditional cloud-only AI models introduce valid concerns around data residency, explainability and third-party exposure.

Experts agree that a sovereign, hybrid AI architecture is a strategic response. Under this model, sensitive and personally identifiable data is processed locally within controlled environments, while cloud AI is applied selectively to non-sensitive workloads. Outputs are merged within governed, auditable frameworks that preserve traceability, model version control and regulatory compliance.

This approach goes beyond risk mitigation. It demonstrates privacy by design, reduces unnecessary vendor exposure and enables AI-powered innovation in claims without compromising data sovereignty. In this context, sovereign AI can act as an innovation accelerator by enabling trusted experimentation with sensitive data, allowing insurers to launch proofs of concept, address diverse workloads and scale new use cases with greater confidence.
Delivering on these ambitions, however, ultimately depends on the strength of the underlying technology foundation.

3. Legacy modernization is no longer optional

AI performance depends on architectural readiness. Insurers continue to cite legacy systems as a major barrier to digitization. Fragmented data environments and rigid core systems limit how effectively AI can scale.

While AI does not eliminate structural complexity, it does expose it. Insurers now face a deliberate strategic choice: enhance stable systems that can support AI extensions, fully modernize where technical debt limits agility, or adopt a hybrid approach that introduces flexibility at the edges while stabilizing core platforms.

Layering AI onto inflexible infrastructure may deliver incremental efficiencies. Aligning AI ambition with a clear modernization strategy creates structural advantage. Modernization is no longer discretionary. It is foundational to scalable, enterprise-wide AI.

4. True ROI comes from end-to-end integration

Even with the right operating model and technology foundation, value is realized only when AI is integrated across the enterprise. Insurance workflows operate as interconnected systems. When AI accelerates only one stage, it can create new and costly bottlenecks.

For example, automating claims intake without redesigning adjudication limits impact. Improving operational design without integrating testing and deployment creates downstream friction. Partial adoption can improve individual tasks, but it cannot transform the operating model.
Real ROI requires embedding AI across the full value chain, preserving context across workflows and integrating orchestration so gains compound rather than fragment.

A common risk is implementing AI in isolation. Enterprise value emerges when AI is integrated across processes, rather than inserted into discrete steps. Achieving this level of integration, however, requires disciplined execution.

5. The AI launch pad: Disciplined scaling with governance

Rather than pursuing broad, simultaneous change, insurers should focus on disciplined, outcome-driven execution built around a few priorities:

  • Identify one or two high-impact use cases tied directly to customer or operational outcomes
  • Define measurable ROI and clear success criteria
  • Assess data, governance and architectural readiness to support scale
  • Build internal capability and education alongside technical deployment

As AI initiatives scale, governance moves from a safeguard to a strategic enabler. Human-in-the-loop and human-on-the-loop models ensure accountability, explainability and regulatory confidence, particularly in customer-facing and agentic use cases.

Responsible AI is not about removing human judgment, but redefining collaboration between people and intelligent systems. Early discipline determines long-term scalability. AI maturity builds over time when initiatives are intentional, time-boxed and grounded in governance.

Translating AI strategy into business value

AI investment is accelerating, but competitive advantage will be defined by how effectively it is scaled. Building what’s next requires enterprise alignment, sovereign and privacy-first architecture, deliberate modernization and true end-to-end transformation, supported by a human and an AI operating model grounded in governance and measurable outcomes.

At CGI, we partner with insurers to modernize core platforms, strengthen data foundations and scale responsible AI across the enterprise. Connect with us to explore how we can help translate your AI priorities into measurable outcomes.