Yesterday Sahar Dolev - Blitental, Flexor’s CMO, hosted the panel “AI in Insurance: From Underwriting to Claims, Where Is It Actually Working?” at CDAO FSI UK, with Darwin Castaneda, MBA (Liberty Mutual), Beatrice Russell, ACII (AON), Alan Strange (Sophro), and Varalakshmi Venkatraman (Marshmallow).
It was a pleasure to hear from the speakers’ experience about how to move AI beyond potential to real value.
A few insights that stood out:
🎯 In underwriting, AI can organize submission data, assess risks against appetite and help underwriters focus on the cases that require closer attention. But assistance isn’t the same as decision-making.
🎯 In claims, AI is helping teams bring together fragmented data, identify potential fraud, prioritize workloads and automate routine processing. People should stay involved where judgment or authorization is required.
🎯 Organizations can achieve meaningful results with very different budgets, as long as the business problem is clear.
🎯 AI is only as useful as the information it can access. Trusted, connected data, explainable data is the way to achieve reliable and explainable outcomes.
🎯 Governance has to work in practice, not just on paper. That means evaluation frameworks, clear guardrails and ongoing feedback from the people using these systems.
And most of these use cases run with unstructured data at the core.
Thank you to our panelists for sharing such practical and honest perspectives and to everyone who joined the conversation.
Corinium Global Intelligence Kathleen Vitug Lauren Hughes