AI in Insurance is a weekly roundup of how artificial intelligence is reshaping the (re)insurance industry—from underwriting and claims to regulation, climate modelling, fraud detection, and beyond. Drawing on industry news from global sources, this blog curates the most important stories and adds a layer of commentary to explore not just what’s happening, but what it means. As insurers move from experimentation to implementation, my goal is to track that shift and spotlight both breakthroughs and blind spots.

This edition captures a clear theme: insurers worldwide are scrambling to keep pace with the rapid rise of AI—and the risks that come with it. Regulatory tensions are mounting on both sides of the Atlantic, customer trust gaps are emerging, and the AI arms race is driving new product innovation, legal challenges, and infrastructure strain. As AI shifts from hype to daily operations, insurers face growing pressure to convert potential into practical protections.

Tech meets captives in a big way here. Parametrics are finally going mainstream—with AI in the mix.

This is the kind of story that sneaks up on you—a decade-long freeze on AI oversight buried in a tax bill? The insurance industry’s backlash shows how high the stakes are for balancing innovation with consumer protection.

This survey confirms what most of us suspect: everyone’s dabbling in AI, but very few have a coherent strategy.

There’s a clear trust gap emerging—AI-savvy customers are warming to automation, but for everyone else, human oversight and clear rules still matter. Insurers chasing efficiency need to remember: confidence builds slower than algorithms do.

Agentic AI isn’t a distant promise—it’s landing in claims today. Amazon’s vision of AI coworkers is already playing out at Agentech, where digital teammates are reshaping claims handling, boosting scale, and giving adjusters time to focus on the tough, human moments that tech can’t replace.

Another clear signal that AI spend is no longer experimental—it’s becoming foundational for underwriting and risk decisions. But the adoption gap between big insurers and smaller players could widen fast if governance doesn’t catch up.

A new underwriting platform focused on AI litigation? If this catches on, it could redefine risk modelling for tech companies. I saw Testudo present at Lloyd’s recently—their approach feels like one of the more practical, near-term plays for tackling AI liability.

This one’s all about ROI. Finally, fraud detection that actually saves real money—Deloitte’s numbers are eye-popping.

AI risks aren’t just theoretical anymore—this shows how edge cases and hallucinations could trigger real-world claims and legal headaches, especially as AI moves from software into cars, homes, and critical systems.

UK risk managers are sounding the alarm on AI, cyber, and geopolitics—but they’re also getting creative with captives and alternative risk tools. The message is clear: traditional insurance is struggling to keep pace with tech-driven business risks.

Insurance has a cyber bullseye on its back. This is your wake-up call from Google’s threat intel team.

It’s a bold move by Reddit, and I think it signals the start of serious pushback from data holders. This could set a precedent.

This is a hidden cost of AI that no one wants to talk about. Insurers should be paying more attention to the environmental backlash.

Finally, some real ROI from AI in insurance. Fraud detection that actually saves money and works faster is worth celebrating.

I’ve seen too many bland proposals—this could change the game for brokers if it lives up to the hype.

A reminder that the AI race is also political. Regulation battles like this will shape how we deploy AI in real businesses.

Underwriters need to stop seeing GenAI as a threat and start seeing it as a tool. This paper might help.

Hiscox is ahead of the curve here. Clearer wording for AI-related risks is exactly what clients need.

Lots of feel-good quotes here, but I’d like to see more hard data. Still, the momentum is undeniable.

Claims is the obvious entry point for AI. It’s messy, repetitive, and full of inefficiencies—prime territory for automation.

Some of this is philosophical, but the point is right: AI will force us to rethink risk from the ground up.

I like the gardening analogy. Without good data, AI is just window dressing. This one’s for the ops folks.