Weecover, an InsurTech platform that builds insurance management and distribution technology, said the real barrier to scaling AI across the sector is not algorithms but the rigid operational architecture inside traditional carriers. The statement lands during a period of concentrated AI investment: Gallagher Re's Global InsurTech Report for Q1 2026 shows 95.2% of the $1.63 billion flowing into the sector went to AI-focused companies. Despite that capital surge, the industry faces a widening gap between proof-of-concept experiments and profitable, large-scale deployment.
The architecture bottleneck
Weecover's core argument is that AI adoption stalls because legacy core systems cannot supply the clean, accessible data flows that machine learning models require. Jordi Pagès, CEO of Weecover, said, "The real bottleneck to AI adoption is no longer the technology itself, but the operational architecture that supports it." The company points to a structural problem: many insurers are attempting to build AI capabilities directly on top of complex, often decades-old infrastructure.
Rather than ripping out those systems, Weecover advocates a modular transition. The approach layers cloud-based solutions built on APIs and microservices alongside existing platforms. This lets carriers introduce new capabilities incrementally without disrupting live operations. A modular middleware layer acts as a translator between AI tools and legacy code, stitching together data that would otherwise remain fragmented across disconnected environments.
Progressive modernization over wholesale replacement
Pagès said, "Rather than replacing the entire core system overnight, insurers can adopt architectures that allow them to modernise progressively while continuing to operate their existing systems." The company, which began as an embedded insurance distribution solution and evolved into a modular SaaS platform for MGAs and insurers, argues that wholesale core replacement is both costly and risky. Its model instead relies on a gradual build-out that keeps current systems running.
The technical foundation for this approach rests on APIs and microservices that let AI, automation, and advanced analytics access and share data across the full technology stack. Without that data mobility, Pagès said, "innovation is simply not viable."
Why this matters for insurance professionals
For underwriting, claims, and operations leads who are being asked to deliver AI results, the message is practical: the fastest path to production may not be a new algorithm but a data integration layer that sits on top of what already exists. The investment figures confirm that AI is where the money is going. The architecture question determines whether that money produces a return. Professionals evaluating AI for Insurance initiatives should scrutinize whether their current infrastructure can feed models the consistent, high-quality data they need, or whether a modular middleware investment must come first.
Your membership also unlocks:
