OpenText Redefines Enterprise Automation And AI

OpenText World 2025 brought together practitioners and experts from across the technology landscape to explore one defining question: Can AI truly transform the enterprise in an era where trust, context, and security are non-negotiable? This collection of analyst perspectives examines OpenText’s latest announcements and innovations.

Security Underpins AI Initiatives But Was Absent From The Mainstage

Joseph Blankenship, VP, Research Director Despite mentioning that security and trust are foundational for all AI initiatives, security was largely absent from the mainstage keynotes. There were occasional mentions of security but even the session led by Muhi Majzoub, EVP of Security Products, focused on a client’s AI innovation, not security. Despite their absence from the main stage, application security and identity were included in breakout sessions. Threat detection and response and security operations, however, were noticeably absent from the event. Perhaps this signals their diminished importance as part of the product portfolio as OpenText focuses on capabilities to enable customers’ AI initiatives. OpenText emphasized the need for guardrails and observability to secure AI. It also highlighted the need for data security and managing agent identities throughout the event, noting its data security and IAM capabilities. The vendor introduced its “Trust By Design” model for securing AI agents, which includes identifying sensitive data, securing identities, monitoring behavior, securing apps, and being ready to respond (see Forrester’s AEGIS model for more on securing agentic AI). OpenText also showcased its SAST Aviator, an AI agent designed to reduce false positives in application testing and provide remediation advice. SAST Aviator must be invoked after the SAST scan is complete. Customers can choose from three deployment models: On-premises, SaaS, and single-tenant hosted. SAST Aviator uses AI to determine whether identified issues are true or false positives, enabling human-in-the-loop review and automated remediation. OpenText’s SAST is one of the furthest along in being able to identify OWASP Top LLM security flaws through static analysis such as prompt injection, excessive agency, and encoding confusion (invisible characters) by investing in languages and frameworks common to AI applications. To help enterprises move to a post-quantum future, OpenText highlighted several key capabilities. OpenText SAST will identify non-PQC resilient algorithms (RSA and DSA) in Java code bases and OpenText DAST will detect Missing PQC Resilient Key Exchange. OpenText plans to expand support to more quantum-vulnerable algorithms, analyze cryptographic configurations for post-quantum security, and extend detection capabilities across multiple programming languages.

Peace, Order, Good Government – and AI?

Andrew Cornwall, Senior Analyst Canadians have an expression that exemplifies Waterloo, Ontario-based OpenText: Peace, order, and good government. In the past, OpenText has focused on content and document management with the auditability needed in highly regulated industries. As AI and agents increasingly become talking points for CTOs at Fortune 100 companies, OpenText has pivoted to emphasize AI informed by trustworthy data that’s guarded by robust access controls. With that ambition comes a strong interest in context and trustworthy AI. OpenText’s opinion appears to be that, with enough context, AI can be made 100% correct. I question that: As long as language models underlie AI, hallucinations will remain statistical probabilities. OpenText’s DevOps platform has always emphasized its testing strengths over build or deploy. QA teams at large enterprises have relied on it for self-healing functional testing (UFT, now called OpenText Functional Testing) and performance testing (LoadRunner, now called OpenText Performance Engineering). But OpenText’s AI for DevOps feels as if it’s a little behind OpenText’s AI for everything else. Yes, there’s the new Aviator EvalOps, a solution for evaluating and benchmarking AI in applications. However, it wasn’t clear to me whether EvalOps is part of the new OpenText Aviator Studio, an AI control plane, or included in OpenText Software Delivery (née ValueEdge). Functional Testing sees Python support, and there’s an MCP server for Performance Engineering. OpenText has said they’re committed to vendor neutrality in tooling and they continue that with language models, although there was mention of validation requirements before you could use an arbitrary model. OpenText demonstrated their AI integration by turning product requirements into work items, which had some integration but included at least a few awkward copy/paste interactions. As a company, it’s not clear that OpenText understands the needs of developers, as opposed to those of QA. At least two OpenText employees proudly told me that, “We integrate with Jira because developers love their Jira.” Perhaps Canadian developers are being polite.

From Content In Context To AI In Context

Cheryl McKinnon, Principal Analyst OpenText has helped organizations manage and protect their content for more than 30 years. This content is now a major pillar in OpenText’s new positioning as “Secure Information Management for AI”. Well-governed content is already rich with context – metadata, audit trails, workflows, permissions, folders, and workspaces – but OpenText recognizes that the new era of AI agents needs a deeper layer of insights. The announcement of the OpenText AI Data Platform was the top announcement of the event. OpenText Content Management customers will be able to take advantage of this cross-product line and unified stack as of mid-2026 with the 26.2 release, and have enhancements extending into 2027. Key components of the AI data platform include a knowledge graph to visualize relationships across structured and unstructured data, integrated applications such as SAP, an AI Control Plane, Knowledge Discovery (powered by IDOL), and data governance to assure security and compliance. Content Aviator customers will be able to take advantage of the OpenText Aviator Studio tool by mid-2026, allowing knowledge workers to build, test, and manage their own AI agents to perform specific actions with content, processes, and related data. A free tier of Content Aviator will be bundled into private cloud “X” plan subscriptions as of release 26.1, enabling search and summarization for all users. (This is already available in OpenText’s SaaS content platform, Core Content Management). But to get to this next level of AI and innovation, many customers need to act. Customers sitting on older versions of OpenText Content Management (formerly known as Extended ECM) or Documentum will need to modernize and get to current software. To help these clients, OpenText also announced OT Migrate, a productized migration tool to fast-track upgrades and move to the OpenText private cloud. A defined set of packaged professional services offerings will supplement the migration tools, helping legacy customers optimize their implementations (such as adopting contextual business workspaces).

Intelligent Service And Knowledge Transformation For The Future

Julie Mohr, Principal Analyst OpenText’s recent announcements signal a strong push toward intelligent enterprise automation and knowledge-centric operations, reflecting a broader industry shift toward smarter, more adaptive IT ecosystems. The updates emphasize streamlined service delivery, improved user experience, and deeper integration of AI-driven insights into everyday workflows. One of the most notable strengths is the commitment to embedding artificial intelligence into operational processes. By leveraging predictive analytics and generative capabilities, OpenText aims to accelerate issue resolution and reduce manual intervention. Global Mining reported, “Faster decisions mean safer operations, which means a new frontier of exploration powered by insight and not guesswork.” Thales shared, “The result was an incredible 85% reduction in alert noise. ​Analysts could now focus on what mattered most, identifying real threats, responding faster, and fortifying the very systems that protect lives and nations.” The ability to dynamically surface relevant information and continuously refine knowledge assets based on real-time data positions OpenText as a forward-thinking player in enterprise automation. However, these advancements are not without challenges. Success depends heavily on the maturity of AI models and the quality of underlying data. Zurich Airport stated, “Today’s data at the airport isn’t noise, it’s insight. The team can see patterns before problems. ​They can act, not react.” Organizations with fragmented systems or legacy environments may face hurdles in realizing the full potential of these features. Meanwhile, integration complexity is another concern as embedding advanced capabilities into diverse IT landscapes often requires significant investment and planning. In essence, OpenText’s strategy reflects a bold vision for the future of IT operations: One where enterprise automation and knowledge work hand in hand to deliver efficiency and resilience. Yet, the journey from promise to performance will hinge on governance, execution, and the ability to translate innovation into real-world outcomes.