The Gap Between AI Pilot and Production Governance
Calling a language model is easy. Running it under enterprise control is not.
The moment an agent needs live enterprise data, permission to act, and an accountable record of what it did, most teams discover their AI stack sits outside every control the business runs on. Credentials spread across cloud runtimes. Data gets copied to wherever the AI lives. Each new agent becomes a fresh security review and risk.
InterSystems IRIS® AI Hub closes that gap by making the data platform itself the place where AI is built, run and governed, instead of bolting another stack alongside it. Teams build native AI agents, embed AI in business workflows with a human in the loop, publish business logic as governed tools using Model Context Protocol (MCP), and connect AI to live enterprise data — whether that data already sits in InterSystems IRIS or across the systems InterSystems IRIS integrates with.
The moment an agent needs live enterprise data, permission to act, and an accountable record of what it did, most teams discover their AI stack sits outside every control the business runs on. Credentials spread across cloud runtimes. Data gets copied to wherever the AI lives. Each new agent becomes a fresh security review and risk.
InterSystems IRIS® AI Hub closes that gap by making the data platform itself the place where AI is built, run and governed, instead of bolting another stack alongside it. Teams build native AI agents, embed AI in business workflows with a human in the loop, publish business logic as governed tools using Model Context Protocol (MCP), and connect AI to live enterprise data — whether that data already sits in InterSystems IRIS or across the systems InterSystems IRIS integrates with.
AI Value Starts with a Governed 360° View
Business Data
Direct access to the most relevant information is critical for Generative AI and Agentic applications to provide relevant information. Tightly integrated with InterSystems IRIS Vector Search and running right where your data lives, the AI Hub offers that complete picture, without the hassle of copies, connections, and caches.
Business Applications
To make Agents act rather than just answer, they need access to tools that are relevant to the current context. Agents running in the AI Hub can access native tools directly, and exposing InterSystems IRIS business logic as MCP Tools takes just a few clicks. Use your existing code without writing anything new.
Business Security
Agents have a role just like human rules, and their access to resources needs to be governed, so why would you manage those identities and permissions in a different system? InterSystems IRIS AI Hub lets you reuse your existing roles and access policies, such that you can govern your agentic and human users in a common language.
Business Workflow
In a real-world business process, approvals go beyond chatbot elicitations, prompting someone looking at the screen right then and there. InterSystems IRIS AI Hub integrates seamlessly with InterSystems IRIS Interoperability, including its extensive human workflow facilities to set up advanced approval workflows.
The InterSystems Difference
Agents Run Inside the System of Record
Building AI inside the platform that holds live transactional data removes an entire class of work: no export, no replica, no drift between the copy the model sees and the record the business runs on.
Governance and Security From Build Time to Run Time
InterSystems IRIS AI Hub was built from the ground up, deeply integrated with the platform. No outside dependencies means a more secure software supply chain, and the ability to express security policies for agents just like your existing business users.
Industry-Leading Support
We're focused on making our customers successful and ready for any challenge, which is demonstrated by some of the highest customer satisfaction ratings in the category.
Powering the World’s Most Important Applications
Our software powers mission-critical applications in nearly every industry — from healthcare and financial services to supply chain and space exploration.
AI That Fits the Way You Already Work
Build AI using design patterns you already know, connect it to the tools your teams already use, and govern it with the controls you already run.
Build
Create agents, assistants, and retrieval solutions with native InterSystems IRIS classes or Python. InterSystems IRIS developers work in the patterns they already know; Python and Java teams keep their familiar tools while drawing configuration and credentials from the platform.
Connect
Connectivity via native, bidirectional Model Context Protocol. Publish governed tools to the AI clients your teams already use, and let native InterSystems IRIS agents consume external tools and data sources in turn. Neither side is a closed ecosystem.
Ground
Anchor answers in enterprise reality using InterSystems IRIS Vector Search with your live operational, clinical, and transactional data — not a snapshot taken when the pipeline last ran.
Govern
Enforce role-based access to models and tools, hold credentials in the InterSystems IRIS Wallet, and manage configuration centrally rather than per project in a separate stack
Add Governed AI to Your Critical Data
The AI Hub adds governed AI to InterSystems IRIS if you're already running it, or serves as the governed layer between your AI and the systems you already operate if you're not. Either way, agents reach the data they need without it being copied into a separate AI platform. The business logic, data model, and permissions you've already built become the substrate for AI—rather than something AI works around.
Resources and Documentation
Aug 14, 2026
Video
A three-minute overview of the key capabilities offered by the InterSystems IRIS AI Hub.
Aug 10, 2026
Product Data Sheet
A two-page summary of what the AI Hub does, who it is for, and what it replaces.
Documentation
Consult the full product documentation and start building today.
May 08, 2026
Keynote Presentation
InterSystems IRIS® AI Hub announcement at InterSystems READY 2026
See What Governed AI Looks Like on Your Data
We’d love to talk. Fill in some details and we’ll be in touch.