Leveraging AI in Insurance: Starting Small with Strategic Vision for Transformative Impact

Executive Summary

The adoption of artificial intelligence (AI) is reshaping industries worldwide, and the insurance sector stands to benefit significantly by following a measured yet visionary approach. Insights from Vivian Sun, senior director of Data & AI at Jabil, as featured in a Wharton discussion, underscore the importance of initiating AI projects with clear business value and scalability in mind rather than technology for technology’s sake. Jabil’s experience in deploying AI for quality control and trade compliance exemplifies how insurance companies can harness AI to improve operational efficiency, enhance risk assessment, and elevate customer service.

For insurance professionals, this means embracing AI not as a replacement for human expertise but as a powerful augmentation tool that can handle repetitive tasks, support complex decision-making, and reduce errors. Starting with small, manageable AI initiatives that directly address pressing business challenges allows insurers to build confidence, demonstrate ROI, and pave the way for broader digital transformation. The evolving role of AI agents as digital coworkers also invites insurance firms to rethink workforce strategies and regulatory compliance frameworks.

Key Insights

1. Business-Driven AI Implementation
   AI projects should originate from specific business problems rather than the allure of emerging technologies. Jabil’s approach focused on solving practical issues, such as color matching and defect detection, ensuring AI delivered tangible improvements. For insurers, this means identifying pain points like claims fraud detection, underwriting bottlenecks, or policy servicing inefficiencies as starting points for AI initiatives.

2. Scalability and Long-Term Vision
   Early AI successes must be designed with scalability to enable wider organizational transformation. Jabil’s initial AI use for color accuracy expanded into computer vision for defect inspection, demonstrating the value of building on foundational wins. Insurance companies can similarly pilot AI in targeted areas before scaling across underwriting, claims processing, and customer engagement.

3. Human-AI Collaboration
   AI is best positioned to augment human workers by automating repetitive tasks and providing decision support, not to replace them entirely. Jabil’s experience relieving employees from monotonous inspections without job cuts illustrates how AI can improve job quality and productivity. In insurance, AI can assist agents and underwriters by flagging anomalies, suggesting risk scores, or automating routine communications, thereby enabling staff to focus on complex, value-added activities.

4. Layered AI Approaches for Accuracy
   Combining multiple AI methods—such as machine learning with generative AI, can enhance reliability in critical processes. Jabil’s dual-layer validation for international trade codes reduced errors by cross-checking outputs from different AI models. Insurers could apply this layered methodology in fraud detection or regulatory compliance to minimize false positives and ensure accuracy.

5. Emergence of AI Agents as Digital Coworkers
   The concept of “AI agents” that operate with human-like agency and align with company policies suggests a future where AI entities participate actively in workflows. Insurance firms may soon deploy AI agents to handle customer inquiries, process claims, or assist in underwriting decisions, necessitating new governance models and training frameworks.

Insurance Industry Applications

• Claims Processing Efficiency:
  Insurers can implement AI-powered computer vision to automate damage assessments from images, similar to Jabil’s use of AI for visual defect detection. This reduces turnaround times and improves accuracy in claims adjustment.

• Underwriting Augmentation:
  AI can replicate the tacit knowledge of experienced underwriters by analyzing vast datasets and environmental factors to recommend risk ratings. Layered AI systems can validate risk assessments, decreasing underwriting errors and speeding policy issuance.

• Fraud Detection and Prevention:
  By employing machine learning models enhanced with generative AI chatbots, insurers can create a two-tier fraud detection system that cross-verifies suspicious claims data, reducing false alarms and improving investigative focus.

• Regulatory Compliance:
  AI agents can be trained to monitor and interpret evolving insurance regulations, ensuring accurate coding and documentation, much like Jabil’s approach to managing complex trade codes. This reduces compliance risk and administrative burden.

• Customer Service Enhancement:
  Deploying AI agents as first responders to policyholder inquiries can provide consistent, policy-aligned answers quickly, freeing human agents to handle more nuanced interactions and improve customer satisfaction.

Conclusion and Recommendations

For insurance professionals aiming to unlock AI’s potential, the key takeaway from Jabil’s journey is to start small but think big. Focus initial projects on well-defined business challenges that promise measurable improvements. Prioritize scalable solutions that can evolve into comprehensive digital transformation initiatives. Embrace AI as a tool to empower employees rather than replace them, fostering collaboration between human expertise and machine intelligence.

Moreover, consider implementing layered AI approaches to enhance accuracy and reliability in critical functions such as claims adjudication and compliance. Prepare for the integration of AI agents into workflows by establishing robust training, validation, and governance frameworks aligned with regulatory requirements.

By following these principles, insurance companies, agents, and underwriters can navigate the evolving AI landscape confidently, driving innovation and sustained competitive advantage.

For a deeper exploration of these concepts, see the original discussion with Vivian Sun at Wharton’s Knowledge site: Adopting AI: Starting Small While Thinking Big.