Perspectives / Insurance & AI
Beyond the AI pilot: redesigning the work.
McKinsey’s AI research points to a widening gap between adopting AI and changing how an enterprise actually operates. For insurance, that gap is where the most important work begins.
A Loresse perspective · Based on executive briefings prepared in June 2026
The central idea
Adoption is not the same as impact.
The briefing materials behind this piece interpret McKinsey’s AI research through an insurance lens. Their central reading is straightforward: using AI broadly does not, by itself, produce enterprise-wide change. The work, the data that supports it, and the way people supervise it all have to change together.
In an insurance operation, a claim can move between intake, triage, coverage review, documentation, settlement, payment, and audit. Each handoff may involve different systems, people, and rules. Improving one isolated task leaves much of the surrounding friction intact.
From research to operating practice
Four shifts that matter.
Redesign the workflow
Adding an AI assistant to a broken handoff does not fix the handoff. The more consequential question is how intake, review, exceptions, and approvals should work together from start to finish.
Connect the context
Policies, claims notes, forms, images, correspondence, and system records all contribute to a decision. AI needs governed access to that context, with sources and reasoning a specialist can inspect.
Keep judgment accountable
Agents can assemble evidence, recommend next steps, and surface exceptions. Experienced people remain responsible for interpreting ambiguity and approving consequential decisions.
Make oversight part of the work
Permissions, traceability, and review should travel with the workflow itself, rather than being added after an automated decision has already been made.
The insurance opportunity
An operating layer, not another isolated tool.
Carriers, MGAs, and third-party administrators work across fragmented records and highly specific requirements. A connected workflow can bring relevant context to the right person, route exceptions for review, and retain a traceable history of what happened and why.
For Loresse, the aim is to turn institutional knowledge and specialists’ judgment into reviewable AI-enabled workflows. A focused starting point, such as claims intake or liability review, can establish the evidence and oversight needed to consider adjacent steps. That is an expansion path to validate, not a claim that every workflow is already deployed or every outcome is guaranteed.
About this perspective
Sources & attribution.
This is Loresse’s synthesis of the executive briefing materials prepared for Loresse in June 2026. Those materials interpret McKinsey AI research alongside Loresse’s own strategic positioning. The observations about Loresse and the insurance applications described are Loresse’s interpretation, not McKinsey findings, endorsements, or verified production outcomes.
Start a conversation