Artificial intelligence (AI) has rapidly shifted from promise to priority in the P&C insurance sector. With rising pressure on margins, increasing operational complexity, and growing customer expectations, insurers are turning to AI to improve speed, accuracy, and efficiency across the insurance lifecycle.
The momentum is clear: more than 60% of P&C insurers are piloting or deploying AI. Yet fewer than 15% have scaled AI across core operations, and only about a quarter feel confident in their organization’s AI maturity. The result? Most insurers remain stuck in pilot mode, experimenting with promising tools but struggling to realize the enterprise-level value they promise.
Insights from the AWS GenAI Bootcamp at Guidewire’s Annual Conference reflect this reality: only 20% to 30% of generative AI pilots in insurance make it into production. Unlocking that value requires more than enthusiasm. It demands a purposeful approach to execution—one grounded in strategy, infrastructure, and measurable outcomes.
From Pilots to Progress: Addressing the Challenges That Hold Insurers Back
Despite increased investment and broad experimentation, most AI initiatives in the insurance industry still struggle to scale.
The reasons are well understood: Fragmented or siloed data often prevents organizations from training robust models or generating useful insights. Legacy IT systems create integration roadblocks, making it difficult to embed AI tools into real-time workflows. Even when pilots show promise, unclear or inconsistently tracked metrics make it hard to quantify results and justify further investment. And without strong governance frameworks—including oversight of model risk, accuracy, and compliance—many insurers hesitate to expand AI beyond small-scale tests.
These constraints prevent even high-potential pilots from advancing and keep AI from becoming the transformative business accelerator it promises to be. To move beyond the pilot stage, insurers must address these foundational roadblocks head-on.
A major contributing factor is the disconnect between teams: business leaders, data scientists, and IT departments often operate in silos. As a result, the capabilities of AI tools can diverge from the practical needs of those on the front lines.
In short, without the right focus, infrastructure, cross-functional alignment, and governance in place, even the most innovative AI use case risks stalling before it can deliver enterprise value.
Unlocking Value: What AI Success Looks Like
Insurers that are scaling AI effectively tend to share three defining characteristics:
First, they target and focus on high-impact business problems that are directly tied to operational pain points—whether it’s accelerating submissions, streamlining claims triage, or improving pricing accuracy. These aren’t experimental side projects; they are targeted initiatives aimed at real inefficiencies that affect business performance. When these efforts deliver quick wins, they build internal confidence, strengthen stakeholder buy-in, and create momentum for scaling AI more broadly across the organization.
Second, successful insurers operate on modern, scalable infrastructure capable of supporting real-time data processing and model deployment. Without a flexible and resilient technical foundation, even the most promising AI models can falter before they reach production.
Third, they build in governance and metrics from the start. Structured testing environments, model monitoring, and clear accountability frameworks ensure that AI performance is transparent, responsible, and aligned with both business objectives and regulatory requirements.
In short, AI success is driven by intentionality, not novelty. It’s not about chasing the latest trend or deploying flashy tools—it’s about identifying real business problems, applying the right technologies to solve them, and embedding those solutions into the day-to-day fabric of operations. The insurers making progress aren’t experimenting for experimentation’s sake—they’re setting clear goals, aligning stakeholders, and focusing on measurable outcomes. That discipline is what separates organizations stuck in pilot mode from those unlocking real enterprise value.
These principles are playing out in real-world deployments. In underwriting, AI is being used to predict risk and refine pricing using behavioral data, location-based insights, and third-party sources, often embedded directly into systems like Guidewire PolicyCenter to support smarter decision-making at the point of quote. In claims, machine learning models embedded in platforms like ClaimCenter are detecting fraud more accurately, improving triage speed, and reducing loss adjustment expenses. Generative AI is further expanding its impact by enhancing customer experience, powering virtual assistants, summarizing lengthy call transcripts, and automating policy documents and email responses.
Infrastructure for Impact: Operationalizing AI with AWS and Guidewire
AI success doesn’t come from experimentation alone; it comes from operationalization. That means embedding AI where decisions are made and ensuring it actively supports day-to-day workflows across underwriting, pricing, claims, and customer engagement.
This is where Guidewire Cloud, built on AWS, provides a strong, scalable foundation. Together, they enable insurers to identify services that can deploy both predictive and generative AI capabilities inside the systems their teams already use.
Guidewire’s core platforms—PolicyCenter, ClaimCenter, and BillingCenter—are becoming the natural homes for AI innovation, allowing frontline staff to access AI-powered insights without ever leaving their daily environment.
To support this, AWS offers two powerful—but distinct—AI services:
Amazon Bedrock is AWS’s generative AI service that lets insurers rapidly build and scale applications using foundation models from top providers (e.g., Anthropic, Meta, and Amazon). It’s ideal for automating tasks that involve unstructured data, like summarizing adjuster notes, extracting information from customer emails, or powering intelligent virtual assistants, without managing infrastructure or tuning models from scratch.
Amazon SageMaker is a fully managed machine learning (ML) platform for predictive AI use cases. It supports the full ML lifecycle and can be used by insurers to develop advanced models for risk scoring, pricing optimization, fraud detection, and claims severity forecasting. These models can be integrated directly into Guidewire workflows for real-time insights within ClaimCenter or PolicyCenter via Guidewire Analytics.
In short, insurers should use Amazon Bedrock when they want to automate or generate insights from unstructured text, such as summarizing claim notes or powering virtual assistants. In contrast, Amazon SageMaker is best suited for making predictive decisions based on structured data, such as scoring risk or detecting fraud. Together, they provide complementary paths to intelligent automation and decision support.
These AWS services give insurers a full spectrum of AI capabilities, from rapid deployment of low-code GenAI tools to the creation of custom-built predictive models.
And because they integrate natively with Guidewire Cloud, these capabilities don’t sit on the sidelines; they activate inside critical workflows. AI becomes part of how underwriters evaluate submissions, how adjusters triage claims, and how service teams engage policyholders.
By combining Guidewire’s insurance-specific data models and workflow intelligence with AWS’s enterprise-scale AI infrastructure, insurers gain the flexibility, governance, and technical depth needed to move from pilots to production, and from insight to impact.
Insurance-Ready Apps Accelerate AI Innovation
Once the right foundation is in place—modern infrastructure, integrated workflows, and scalable AI services—insurers can begin to unlock the full potential of AI. That’s when innovation accelerates. Whether the goal is enhancing underwriting precision, streamlining complex claims, or transforming customer engagement, the next step is activating AI across the business in targeted, practical ways.
The Guidewire Marketplace makes this possible. With over 130 insurance-ready extensions for Guidewire, many built on AWS, insurers have fast access to pre-validated solutions that extend the capabilities of Guidewire Cloud. These extensions help fast-track AI adoption in critical areas like claims automation, fraud detection, risk assessment, and document processing.
Each is purpose-built for the complexities of P&C insurance, from nuanced policy structures to regulatory constraints and multi-party claims workflows. They integrate directly into core systems like PolicyCenter and ClaimCenter, enabling insurers to test, deploy, and scale with confidence, without compromising security, compliance, or operational stability.
Among the standout Marketplace partners delivering embedded AI innovation:
- Indico Data extracts structured insights from complex commercial insurance documents.
- CogniSure uses AI to extract data from loss runs and policy documents, reducing manual review time.
- FRISS and Shift Technology apply machine learning to detect fraud in real time.
- CLARA Analytics triages injury claims and flags early intervention opportunities.
- EvolutionIQ analyzes structured and unstructured data to surface high-impact claims and guide interventions.
- Tractable uses computer vision to assess damage from images and accelerate settlements.
By embedding these tools directly into daily workflows, insurers can streamline decision-making, reduce manual effort, and drive better outcomes across the insurance lifecycle.
Together, Guidewire and AWS offer a proven ecosystem that reduces development risk, accelerates time to value, and transforms AI potential into operational advantage.
From Vision to Impact
AI is no longer a future-facing initiative—it’s a present-day imperative. For insurers navigating tightening margins, rising risks, and growing customer expectations, AI offers a powerful lever to drive transformation. But turning AI from potential into performance requires more than experimentation. It demands a strategy rooted in real-impact use cases, a foundation built on scalable infrastructure, and a commitment to operationalizing AI across the insurance lifecycle at the points where professionals can leverage its speed and insights.
The insurers leading this shift aren’t just adopting AI—they’re embedding it into how they underwrite, price, triage, and serve. They’re using AI to reduce friction, unlock insights, and move faster in a fast-changing world.
With the right tools and partners—like the Guidewire Cloud Platform, AWS AI services, and pre-validated extensions from the Guidewire Marketplace—insurers can move beyond pilots and build a repeatable path to innovation. But the journey begins with intention: choosing to scale what works, measure what matters, and reimagine what’s possible.
The future of P&C insurance will belong to those who act. Now is the time to start planning and turning AI into an engine for transformation—one that drives smarter decisions, stronger outcomes, and lasting competitive advantage.
Explore the AI solutions and applications in the Guidewire Marketplace.
Read How Guidewire Marketplace Drives P&C Innovation with Insurtechs
Read about AI applications in claims in this Guidewire blog post.