Why 90% of AI Pilots Stall (And How to Escape Pilot Purgatory)
Why 90% of AI Pilots Stall (And How to Escape Pilot Purgatory)  
Podcast: Full Episodes | Insurtech Leadership Podcast
Published On: Thu Mar 05 2026
Description: Episode Overview How many insurance organizations have launched an AI pilot, watched it work in the lab, and then watched it quietly disappear in production? Today we're dismantling the myth that AI adoption is a technology problem—and making the case that it's a workflow problem. Guest Jake Sloan, VP of Global Insurance at Appian, is an operator first. He has run large-scale insurance operations, owned a $150M P&L, and delivered transformations that only happen when you understand how work actually moves—handoffs, exceptions, controls, and accountability. At Appian, he leads the insurance vertical for a process automation and low-code platform focused on claims, underwriting, and operational process orchestration. Key Topics • Pilot Purgatory: Why 90% of AI Projects Stall — Pilots work in controlled environments. Then reality hits: no data pipeline, no workflow integration, no governance, no frontline buy-in. Organizational alignment—not technology—is the breaking point. • The Orchestration Layer — Appian's core thesis: build an orchestration layer first. It sits between your legacy monolith and the next chapter. Additive, keeps work flowing during transformation, and creates the foundation where AI and automation actually stick. • Email as Infrastructure — Underwriters spend 40%+ of their day in inboxes. The AI mailbox use case embeds AI into a workflow that extracts data, routes work, makes decisions, and triggers actions. Underwriters gain 2–3 hours a day back. • Claims Velocity: Days to Hours — One global insurer went from 24–72 hours (FNOL to assignment) to minutes. Digital intake feeds orchestrated workflows. AI triages, categorizes severity, flags fraud risk. The adjuster gets a complete, pre-organized package. • Alignment = Culture, Not Just Tech — Appian's workshops put business, IT, data, and operations in one room to design the ideal state and work backward. Underwriters don't get replaced—they become superhuman. Admin work gets stripped away. • The Talent Problem Is a Workflow Problem — Entry-level insurance work is repetitive email categorization. When AI handles the mundane, these jobs become analytical and attractive again. The organizations winning reskill existing teams and position domain expertise as more valuable. • The 90-Day Deployment Mindset — Pick one workflow. Build the orchestration layer. Plug in AI. Show ROI in 90 days. Then iterate. Key Quotes -"When you go to scale it, it's like, well, we don't have the data pipeline. We don't have the workflow. We don't have the governance. We don't have the buy-in from the frontline team. And so it just stalls." -"It's not about the AI itself. It's about the workflow that the AI sits inside." -"We're not here to replace underwriters. We're here to make them superhuman." -"Get started. Don't let perfect be the enemy of good. Start with a use case, build the orchestration layer, plug AI into it, show value in 90 days." Resources • Appian: appian.com • Jake Sloan LinkedIn: https://www.linkedin.com/in/jacobpsloan] • Joshua R. Hollander, Host: https://www.linkedin.com/in/joshuarhollander/ • Horton International: https://www.horton-usa.com/ Subscribe & Follow Never miss an episode. Subscribe on Spotify, Apple Podcasts, YouTube. Follow the show on LinkedIn: https://www.linkedin.com/showcase/insurtech-leadership-show #InsurTech #Insurance #InsuranceInnovation #FutureOfInsurance #ExecutiveLeadership Subscribe & Follow Never miss an episode. Subscribe on your favorite podcast platform—Spotify, Apple Podcasts, YouTube. Follow the show on LinkedIn: https://www.linkedin.com/showcase/insurtech-leadership-show #InsurTech #Insurance #InsuranceInnovation #FutureOfInsurance #Leadership #ExecutiveLeadership