Why Most AI Pilots Fail
And How to Make Yours Succeed
Insurance organizations are investing heavily in AI. Pilots are launched, proof-of-concepts are presented, and then — quietly — most of them stall. They don’t fail loudly. They just never make it to production.
Integration complexity, process dependencies, accuracy gaps, and the simple reality that most AI tools require you to change how you work before they can deliver any value — these are the patterns that kill pilots before they scale. The result is a growing gap between what carriers, MGAs, and brokers are promised by AI vendors and what they actually see in operations.
This session examines what separates pilots that die in proof-of-concept from automation that reaches production and delivers measurable results. The conversation will focus on the structural reasons AI initiatives stall, how leading operations teams are approaching automation differently, and what it actually takes to move from a promising demo to a workflow that runs reliably at scale.
What You’ll Learn
- Why most AI pilots stall before production The common structural reasons automation initiatives fail to scale — and how to identify them early.
- What separates demos from deployments How leading operations teams evaluate AI differently, and what due diligence actually looks like in practice.
- How to design for production from day one Building automation around existing processes and tools, without requiring costly integrations or workflow overhauls.
- How to measure what matters The metrics operations leaders should be tracking — and the ones that tend to obscure whether automation is actually working.
Featured Speakers
Sasha Haco
Sasha Haco is Co-Founder and CEO of Unitary, an AI automation company building Virtual Agents that handle complex operational workflows with human-level precision. Before founding Unitary, Sasha completed her PhD in Theoretical Physics at the University of Cambridge, where she worked alongside Stephen Hawking, Malcolm Perry, and Andrew Strominger on the black hole information paradox. Her research on black hole entropy and soft hair was published in the Journal of High Energy Physics and she was featured in the 2020 Netflix documentary Black Holes: The Edge of All We Know. She previously founded Ditto Software in 2018 and was honoured with the Women in Innovation award in 2021. Sasha brings a rare combination of deep technical rigour and operational focus to the challenge of making AI automation reliable enough to trust at scale.
Second Speaker
To be announced — a carrier or broker perspective from the InsurTech NY network.
