Insurance market cycle and AI disruption

We’re seeing it up close: the back office of insurance is being rebuilt by software, not people. Last week, I wrote about what happens when professional services clients stop paying for inefficiency. This is the next chapter, with a closer look at the insurance sector. We’ve looked at nearly a dozen AI startups automating the work that BPOs have handled for years. The picture isn’t simple, but the direction is clear. A quiet shift is underway in insurance distribution. Not at the front end, but in the workflows: quoting, policy checks, certs, submissions, and proposals. For two decades, BPOs like Patra, ResourcePro, and Xceedance scaled by taking that work offshore. They built strong businesses on process depth, labor efficiency, and repeatability. Now AI-native startups are targeting the same functions. They are automating quote comparison, policy checks, and proposal development. This isn’t cheaper labor. It’s no labor. At first glance, it looks like disruption. But the dynamic is more complicated. Three forces are now colliding, with everyone fighting for their scrap of margin: – Brokers looking to scale – BPOs trying to stay relevant – AI vendors aiming to replace manual processes with software No one moves in isolation. Each shift affects the others. Everyone is trying to avoid being commoditized. Brokers are experimenting. BPOs are adjusting. AI companies are moving quickly and aiming high. From where I sit, as a venture investor focused on this space, the pattern is clear: as the cost of operations drops, so does the barrier to entry. What becomes more valuable is not process. It is proximity to the insured. The question isn’t who owns the workflow. It is who owns the customer relationship — the trust, the interface, and the ability to guide decisions. That is where power accumulates. And that is where the next winners will emerge.

AI Will Transform Insurance Before Healthcare Insurance was AI's first industry - we just didn't call it that. Statistical learning and probability theory birthed modern insurance decades ago. Now, AI's evolution threatens every incumbent carrier. Manual claims processing wastes millions in human hours. Teams search databases, verify information, and cross-check data while customers wait weeks for decisions. Worse, human bias corrupts every approval or denial. AI doesn't just speed up claims - it removes human prejudice entirely. By analyzing historical data patterns, AI systems can identify and correct biases that hurt specific groups. Instead of training hundreds of adjusters to change behavior, the correction happens at the system level. But current AI faces a critical limitation: correlation versus causation. While neural networks excel at finding patterns, they struggle with true causation. This creates dangerous edge cases where AI might make meaningless recommendations based on spurious correlations. The real revolution won't come from existing carriers adopting AI. It will come from new AI-native insurance companies built from the ground up. Traditional insurers can't transform their decades of human-centric processes fast enough. The question isn't whether AI will transform insurance - it's whether current insurance companies can survive the transformation. Most won't make it.

Insurance is having its AI Moment. A few weeks ago I shared how AI is transforming work and opening up new opportunities to build. For decades, the insurance claims process has relied on a high volume of human coordination: file intake, damage assessment, policy verification, customer communication, and payout. It’s a deeply operational role, but one that has long been constrained by inconsistent documentation, subjective judgment, and bureaucratic lag. Given how (relatively) easy it is to embed AI into products, insurance claims has emerged as the holy trinity of automation readiness: ✅ Structured Data Flows: policy documents, claim forms, repair estimates ✅ High Repetition: at scale millions of similar claims every year ✅ Clear Cutcomes: approve, deny, pay $ As AI takes the bulk of claims processing, human-labor will be increasingly confined to edge cases: complex liability issues, litigation exposure, or emotionally sensitive claims (e.g., life insurance). Today, we’re meeting builders taking aim at established incumbents, looking to reinvent existing workflows with AI, and even rethinking the role of brokerages altogether. In the post I outlined 3 different areas we’re seeing activity in the insurance space:  ➡️ P&C Customer Communications  ➡️ AI-First Brokerages ➡️ Catastrophe Insurance If you’re building in insurance or have ideas for how to reshape the future of work, I’d love to chat.