CData Software was recognized as a Challenger in the 2026 Gartner® Magic Quadrant™ for Data Integration Tools and is the only vendor named a Challenger in this year's report. We believe the placement reflects our Ability to Execute and Completeness of Vision: one data layer that gives people, applications, and now AI agents access to enterprise data.
For the past two decades, CData has built the data layer for people and applications. Our customers haven't changed, but who consumes their data has: agents are now the newest consumer of enterprise data. Enterprise AI will be won or lost at the data layer. That's the layer we're building, and it's why our AI gateway starts there.
One platform as the data layer foundation for people and AI
Production AI needs three things working as one layer: connectivity for reach and depth, context for accuracy and efficiency, and control for scale and trust. Reaching a system is the start. The difference sits above the connection, in whether the answer is right and whose rules apply. The CData Platform brings the main integration styles replication, from batch and change data capture (CDC) to live access, together with the AI gateway on the same connections, so the same governed access serves a dashboard and an agent.
What we delivered this year
Connectivity
Connectivity is where CData started, and we kept expanding it this year. We added change data capture (CDC) for IBM DB2 and SAP HANA, Parallel Partitioned Reads for high-volume replication, enhanced Open Mirroring support for Microsoft Fabric, MQTT and AMQP support, and pipeline orchestration, API 2.0, and Python in Events for the pipelines built on top. The same connections now reach agents, through listings in the ChatGPT Connectors marketplace and published connectors for Anthropic, OpenAI, and Microsoft Copilot Studio.
Context
Reach isn’t enough without meaning. The Context Engine holds four levels of context. System context: what each connected source means, down to the field. Semantic definitions: the models you already maintain, so Revenue means one thing everywhere. Data modeling: views and cross-source joins shaped once and reused on every request. Company knowledge: the unwritten knowledge that lives in people’s heads, captured from use. When someone asks about “this quarter,” it resolves correctly because the Engine knows the fiscal year starts in February. The Gateway learns from every AI interaction, so context is continuously refreshed and expanded.
Control
Policy is enforced before data reaches a person or an agent, down to the field. Each request resolves to the real caller, so an agent inherits that person’s permissions in the source system, and every prompt, query and result is logged. If something looks wrong, access can be shut off for one user, one connection or the whole account in seconds. CData’s controls also support compliance in regulated environments. In the second quarter, we added Business Associate Agreement (BAA) support for HIPAA, alongside our existing SOC 2 Type II, ISO 27001 and GDPR compliance.
Why the AI gateway starts at the data layer
Many AI gateways control routing, access, and spend. For agents to do real work, they also need to understand the systems they act in: what a field means, whose permissions apply, and what an action will change. That understanding lives in the data layer, so that's where CData Connect AI Gateway starts. It governs every step of a request: the model it routes to, the tools it loads, the identity it carries, the records it accesses, and the actions it takes.
Because replication, live access, and policy share one platform, an agent inherits the same connections and permissions your pipelines already use. And because the gateway sits in the traffic path between agents and your systems, it's the one place where context can be learned continuously, so it improves with every query. It works alongside the LLM gateways, model routers, and identity tools you already run. Read more in our September post, The AI Gateway Has to Start at the Data.
Thank you to our customers and partners
The production workloads you run on CData shape what we build next. You push our products and our team harder than any roadmap could, and we are better for it. Thank you for the trust, and for staying with us all these years. We didn’t get here alone.
Read the full 2026 Gartner® Magic Quadrant™ for Data Integration Tools report.
Gartner disclaimer
Gartner, Magic Quadrant for Data Integration Tools, Thornton Craig, Robert Thanaraj, Sharat Menon, Jenna Goodrich, Nina Showell, November 2, 2026.
GARTNER is a trademark of Gartner, Inc. and/or its affiliates. Magic Quadrant is a trademark of Gartner, Inc. and/or its affiliates.
Gartner does not endorse any vendor, product or service depicted in its research publications, and does not advise technology users to select only those vendors with the highest ratings or other designation. Gartner research publications consist of the opinions of Gartner's research organization and should not be construed as statements of fact. Gartner disclaims all warranties, expressed or implied, with respect to this research, including any warranties of merchantability or fitness for a particular purpose.
