As carriers apply AI more deeply in claims and underwriting processes, where the line sits between what AI should decide and what requires a person to own the outcome is still being settled. Getting it wrong creates real exposure in the form of regulatory scrutiny, a loss of policyholder trust, or both. Further complicating matters, drawing that line is getting harder as AI takes on more of the work newer adjusters and underwriters used to learn from. This on-demand conversation is for claims and underwriting leaders at carriers who want a practical answer. Fiona Gill, Senior Vice President of Client Success at Origami Risk, and Mike Pecak, Senior Vice President of Professional Services of Origami Risk, joined John Weber of AM Best for a discussion about what to ask before asking what the technology can do, how to build accountability into a workflow, and what happens when a weak process gets automated. You’ll leave with a simple test you can run against any workflow you’re already automating. You’ll learn how to: Decide which decisions AI handles and which a person owns. Build accountability into a workflow without creating rubber stamps or bottlenecks. Recognize the warning signs of a process automated before anyone owned the outcome. Measure AI value by results, not by how many people stopped touching a process. Apply a simple test to any claims or underwriting workflow you’re already automating.