Wolters Kluwer builds internal AI platform to ship features faster without cutting corners on governance
Wolters Kluwer announced its AI Center of Excellence and FAB platform - a proprietary system designed to let product teams embed AI into customer-facing tools while maintaining the security and auditability that regulated industries require.
The company has spent over a decade integrating AI into its professional software. Rather than treating AI as an add-on, Wolters Kluwer built FAB to make AI a core part of how products work from the start.
How the platform works
FAB handles the operational overhead that slows down most AI projects. It standardizes tracing, logging, model tuning, and evaluation - the infrastructure work that teams typically bolt on after shipping a feature.
The platform supports multiple models rather than locking teams into a single foundation model. It also grounds AI outputs in Wolters Kluwer's proprietary, expert-curated content rather than relying solely on what a language model generates.
For complex workflows, FAB orchestrates multiple AI agents working together with human oversight built in. Integration with existing enterprise systems happens through a governed gateway that maintains security boundaries.
Organizational structure matters
Wolters Kluwer's technology organization - the Digital eXperience Group - supports five divisions through shared platforms and centers of expertise. Each division has a CTO accountable to business outcomes, creating accountability without slowing decisions.
This structure reduced development cycles and increased the portion of digital revenue that includes AI features.
Real products, real constraints
UpToDate Expert AI in health and CCH Axcess Expert AI in tax are already running on FAB. Both sit inside workflows that professionals use daily, where failures carry real consequences.
Because Wolters Kluwer's platforms are cloud-native and API-first, AI features integrate cleanly rather than existing as separate tools. This preserves the auditability and security that regulated industries demand.
For product teams building AI features, the difference between built-in and bolted-on determines whether governance can scale. Bolted-on features break apart under audit pressure. Built-in ones don't.
What this signals for product development
The approach reflects a practical reality: shipping AI features at speed and maintaining enterprise-grade governance aren't trade-offs if you build the right infrastructure first. Most teams face this choice because they didn't invest in it.
Wolters Kluwer's bet is that proprietary content, workflow expertise, and disciplined platform design matter more than raw model capability. That's a defensible position in markets where customers can't tolerate hallucinations or compliance failures.
Learn more about AI for Product Development and Generative AI and LLM systems.
Your membership also unlocks:
