Crestvale Newsroom
Crestvale Newsroom
Gallagher cuts claims review time by 90% with AI
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
Welcome to the daily audio briefing on AI, automation, and business technology for professional service firm leaders. Today, we're looking at how one global firm cut claim review timed to a fraction of what it used to be, and what that shift signals for every firm building AI into core operations. Gallagher just showed what happens when a large services organization stops experimenting and starts building real AI infrastructure. The change is fast, the impact is firm-wide, and it is a clear sign that the firms treating AI as plumbing, not a gadget, are already pulling ahead. Markets closed higher in the previous session. The SP moved up again and stayed in a steady upward trend. The NASDAQ also closed higher, showing the same positive tone. The 10-year Treasury yield rose, which kept pressure on borrowing costs. Bitcoin ended the session higher as well, adding to a week of steady gains. The overall mood stayed optimistic, but with clear signs that money remains expensive. Gallagher's move is worth paying attention to because it breaks from the usual pattern. Most firms talk about AI pilots. Gallagher built an AI platform with real governance, real data controls, and real workflow integration. And they did it across 60,000 employees. The headline number is simple. Claims review time dropped by as much as 90%. What used to take hours now takes seconds. But the point is not the speed. The point is why the speed is possible. They reorganize the work. Claims handlers now generate clean summaries using the same tools they already use. They built a standard AI platform, so every new use case plugs into the same foundation. And they rebuilt complex document work as lightweight AI apps that match how people already move through approvals. This is the part that matters for firm leaders. They stopped treating every workflow as a fresh project. They built shared infrastructure first, then applied it everywhere. This is where many firms get stuck. They build one-off experiments, they run into data problems, they blame the model. Gallagher avoided that because they built the base layer first. Data standards, permissions, consistent workflows, a platform, not a pile of tools. This matters for every professional services firm because the gap between firms with structured AI platforms and firms that rely on ad hoc usage is about to widen. The first group will automate the high volume tasks quickly. The second group will spend the next few years watching cost pressure rise while workflow delays stay exactly the same. Now, the second major shift today is about legal risk. Class action filings are moving into areas most firms treat as low stakes. Website tracking tools, routine breach response, basic AI deployments. Plaintiffs are using old wiretapping and consumer laws to attack practices that most firms still think of as normal. Courts are letting these cases move forward. That changes the risk profile for every services business with a website, client data, or AI tools. The plaintiff's bar has found a new revenue channel, and lean firms without strong governance are the easiest wins. This is not a tech giant problem. It is a standard tools problem. Firms that tighten their tracking, update their breach playbooks, and apply clear AI governance now will avoid the messy discovery fights that are hitting unprepared organizations. Meanwhile, a massive shift in infrastructure is taking shape behind the scenes. More than$600 billion is flowing into new AI data centers. Hyperscalers are spending because the current grid and hardware cannot support the next wave of AI workloads. This is not an expansion, it is a rebuild. Debt financed campuses, power constraints, and shifting regional availability are already influencing cloud pricing and performance. You may not see the changes yet, but they will show up in renewals and in the fine print around data residency and redundancy. Energy is the bottleneck, capacity is tight, and location is about to matter far more than it did before. For firms running cloud-heavy operations, this means two things. Track where workloads actually run, and expect pricing pressure as infrastructure costs flow downstream. Firms that know their regional footprint will negotiate better than firms that treat the cloud as a black box. Another shift worth watching comes from managed service providers. Mature MSPs are now dropping clients who refuse basic security controls. This is not posturing. Regulators now view MSPs as critical infrastructure. One weak client can expose every other customer on the same stack. That means the days of optional multifactor, casual patching, and untested backups are over. If you rely on an MSP, expect stricter onboarding, mandatory baselines, and real audits. Top MSPs will not carry insecure clients anymore. Organizations that meet the standards will keep the best providers. Those that do not will be pushed out or priced out. Here is what else is worth knowing today. Researchers showed how agents in Vertex AI can behave like insider threats when boundaries are loose. Any firm running autonomous workflows should revisit privileged levels before expanding deployment. Australia's cyber regulators want faster intervention powers for incidents affecting critical infrastructure. The message is simple. Regulators expect real operational readiness, not paperwork. Merker confirmed exposure to the light LLM supply chain compromise. It is another reminder that common open source components can create silent downstream risk. Red Hat reported a widening gap between confidence and actual cloud native security. Teams think they are covered, but runtime controls and configurations suggest otherwise. Here is the takeaway treat AI, security, and cloud decisions as system design, not tool selection, because the firms building the foundation now will own the speed, compliance, and cost advantage for years. If this was useful, follow the Crestvale Newsroom daily podcast so you don't miss tomorrow's briefing. Thanks for listening.