Check out Keypoint Intelligence’s Artificial Intelligence Services!
At this year’s OpenText World 2025 conference, the company’s message was clear: If you want artificial intelligence (AI) to work at scale, you must get serious about how you manage information. Not in the abstract, but as a concrete discipline. Executives Sandy Ono and Savinay Berry outlined a shift in how the company approaches AI, putting information governance at the center and helping companies move from isolated AI pilots to enterprise-scale deployment. OpenText’s central thesis is that this transition requires a new operational discipline they call “secure information management for AI.” AI output is only as good as the data feeding it; therefore, high-quality, governed, and contextual data is a prerequisite for accuracy and relevance.
The OpenText AI & Data Platform
The most significant announcement at the event was the introduction of the OpenText AI & Data Platform (AIDP). Designed to unify data management with AI orchestration, AIDP is positioned to be open to third-party integrations and not just solutions from OpenText. The platform relies on three distinct functional layers:
- Knowledge Discovery: This component focuses on ingesting structured and unstructured data. It automates metadata tagging to create reusable “data products” that serve as the contextual fuel for AI agents.
- Aviator Studio: OpenText introduced this as a no-code “control plane” for the AI ecosystem. It allows organizations to build, govern, and coordinate agents while managing identity, access policies, and entitlements.
- Data Compliance Layer: Security remains a priority with this layer applying Personally Identifiable Information (PII) controls, encryption, tokenization, and threat detection to the data products and the workflows of the agents themselves.
While OpenText is currently using the platform internally to build agentic AI solutions for customers, general availability for customers to build their own agents is expected in the middle of 2026.
The Agentic Future and Interoperability
A recurring theme during OpenText World was the expectation that enterprises will not rely on a single AI model, but on hundreds of specialized agents. To support this fragmented landscape, OpenText emphasized interoperability. The platform is designed to support third-party agents and connect with major ecosystems like those from Microsoft, SAP, Salesforce, and Workday. OpenText also intends to support open standards such as the Model Context Protocol (MCP), an open-source standard for connecting AI applications to external systems, as well as Agent-to-Agent (A2A) communication, which is designed to enable seamless communication and collaboration between AI agents.
Worried about security in this AI goulash? You should be. OpenText noted they will use knowledge graphs—a protective, semantic layer between a customer’s raw data and the AI agents accessing it—rather than direct data exposure to secure MCP components. And to address the “black box” nature of AI, the platform includes an audit trail that records every decision an agent makes, which is vital for compliance reporting.
Notably, OpenText signaled a change in how it engages with clients. With the launch of Aviator AI Services, the company is moving away from pure implementation services toward a more consultative model. This will include using forward-deployed engineers and consultants to help customers identify business problems and build pilots before scaling them on the new platform. In its DevOps practice, OpenText is targeting sectors like financial services, life sciences, and automotive with compliance-embedded workflows and integrations with tools such as GitHub Copilot and application security tooling. The aim is to make DevOps pipelines AI-enhanced and regulatory-friendly rather than forcing customers to choose between speed and control. And while the company is bullish on its cloud offerings, it continues to support cloud, on-prem, and hybrid deployments with customers retaining control over data residency and how data is exposed to AI.
Keypoint Intelligence Opinion
There’s no shortage of AI-powered processing and workflow solutions. But for CIOs, the immediate takeaway isn’t “buy another AI tool.” It’s to ask the important questions: Do we have a coherent information strategy for AI? Can we explain, audit, and control what our AI agents are doing? Are our data, security, and compliance teams working from a shared view of risk and value?
With the OpenText World announcements, the company signals that it plans to be a powerhouse in the next-gen agentic AI wave—and it seems to be doing it right. Orchestration of disparate AI solutions will be key for enterprises, and the company’s embrace of model context protocol (MCP) and agent-to-agent protocol (A2A) will allow it to help customers do just that, while its Aviator AI Services show it can pull it all together.
Stay ahead in the ever-evolving print industry by browsing our Industry Reports page for the latest insights. Log in to the InfoCenter to view research and studies on AI solutions and events through our Production Workflow Advisory Service. Log in to bliQ for product-level research, reports, and specs. Not a subscriber? Contact us for more information.
