Joe Zuk on How Insurers Are Moving AI from Pilot to Production

Moving AI from Pilot to Production: Insights from Joe Zuk

AI in insurance is no longer just a buzzword, it’s becoming an operational necessity. Yet, insurers and MGAs often struggle to move from experimentation to real-world impact. In this episode of Pilot to Production, Joe Zuk, operating partner at Altamont Capital Partners, shares his perspective on AI adoption in insurance, operationalizing AI, and building scalable, human-in-the-loop AI workflows.

AI is reshaping insurance, but the real challenge isn’t the technology itself, it’s getting AI to work in real, workflow-specific contexts. Joe emphasizes that successful AI adoption requires explainable, auditable, and scalable solutions tailored to each insurer’s unique operational fingerprint.

This conversation explores how insurers can turn pilots into production-ready solutions, improve broker workflows, and harness agent-based AI to transform claims and underwriting processes.

Watch the Full Episode Here

Why AI Pilots Struggle in Insurance

Insurers frequently launch AI pilots, only to find limited adoption and measurable impact. According to Joe:

“A lot of people engage in what I like to call ‘pilot theater’…they’re not actually trying to ship a product or solve a specific use case.”

The challenge lies in the operationalization of AI, at StitchStudio, we refer to it as the “last mile.” Pilots can demonstrate potential, but without end-to-end workflow integration, AI fails to scale. For example, in underwriting automation, claims adjustments, or broker workflows, the AI solution must fit seamlessly into existing insurance operations rather than act as a standalone experiment.

What Joe Zuk Is Seeing Across the Industry

Joe shares several key trends and insights from his vantage point:

1. AI for Data Foundations:
   “There’s a great use case to just utilize AI to get that foundational data layer ready…then you can start to do some really interesting things.”
   
   Insurers often have vast amounts of data, but it’s fragmented. AI can help organize and clean data, creating a foundation for operational intelligence.
   
   
2. Human-in-the-Loop Augmentation:
   We don’t believe that AI should replace humans, it should enhance them. Joe Zuk echoes that sentiment 
   “AI is starting to augment the human workforce…using AI for claims adjustment as a co-pilot.”
   
   
3. End-to-End Execution Matters:
   “You need to have a system of execution that’s end-to-end from distribution all the way until you service the customer.”
   
   Insurers should focus on real-world impact across the insurance value chain, not just pilots.
   
   
4. Success Requires Market Understanding:
   “Do the folks really understand their market? Do they bring operational execution? And how do we then use technology and digital enablement to create value?”
   
   Understanding insurance fundamentals remains critical for AI success. That’s why we built StitchStudio specifically for Insurance with Insurance-Specific Ontology.

AI as an Amplifier, Not a Replacement

AI in insurance works best when it amplifies human decision-making rather than attempting full automation. Joe emphasizes:

“…insurance is fundamentally a relationship business…and very much so won’t be replaced.”

Applied AI platforms, like StitchStudio, reinforce this principle by providing agent-based AI workflows, where humans remain central while AI handles repetitive or analytical tasks. This ensures compliance-aware AI deployment and preserves trust in broker, MGA, and TPA operations.

Practical Steps to Move AI from Pilot to Production

Joe outlines actionable steps insurers can take:

1. Define Clear Objectives: Know what problem AI is solving.
   
   
2. Build Foundational Data Layers: Use AI to organize and clean existing data.
   
   
3. Integrate Across Workflows: Embed AI into underwriting, claims, and distribution operations.
   
   
4. Maintain Human-in-the-Loop Oversight: Let AI augment, not replace, human decision-makers.
   
   
5. Measure and Iterate: Focus on outcomes: revenue growth, margin expansion, and improved customer experiences.
   
   

Key Takeaway: AI adoption in insurance succeeds when pilots are operationalized with end-to-end execution, rather than treated as experiments.

How StitchStudio Enables Operational AI

StitchStudio provides insurers with agent-based AI workflows that integrate seamlessly into legacy systems. Our platform offers:

• Contextual intelligence for underwriting and claims
  
  
• Compliance-aware AI for regulatory adherence
  
  

By combining workflow orchestration with human-in-the-loop AI, StitchStudio helps insurers move beyond pilots to measurable business impact.