OpenText's Michael Cybala on Content Aviator, Knowledge Graphs, and AI You Can Trust

As enterprises race to scale AI initiatives, many assume that success comes from giving AI access to more data. But the real differentiator isn't more content. It's better context.

In a recent conversation with Michael Cybala, EVP of Product Management at OpenText, Miller explored why enterprise content is becoming one of the most strategic assets in the AI era and why understanding the relationships behind that content is just as important as the information itself.

Context Is What Makes AI Valuable

Enterprise content extends far beyond documents and files. It includes contracts, emails, engineering drawings, videos, customer communications, and countless forms of unstructured data that support daily business operations.

As Gießwein explained, content alone has limited value. The real advantage comes from understanding its context, including who created it, which business process it supports, what customer it relates to, and how it has influenced business decisions.

For AI, those relationships are essential.

Organizations can easily provide AI with access to vast amounts of information. The greater challenge is ensuring AI can distinguish the right information from the wrong information and deliver responses grounded in current business reality.

More Context Doesn't Always Mean Better Context

Throughout the discussion, Miller challenged a common assumption surrounding enterprise AI.

Simply increasing the amount of available information does not improve AI outcomes. Without trusted business context, AI can generate answers that sound convincing while lacking accuracy.

Grounding AI in business processes, structured enterprise applications, permissions, governance, and metadata creates the foundation for reliable decision-making. Rather than treating context as an enhancement, organizations should view it as a prerequisite for trustworthy AI.

Knowledge Is Built Through Relationships

The conversation also examined one of today's most misunderstood enterprise AI concepts: knowledge graphs.

Rather than serving as repositories for every piece of enterprise information, knowledge graphs create value by connecting information across people, processes, business objects, and decisions.

Building an effective knowledge graph requires shared business semantics, governance, trusted taxonomies, and ongoing maintenance. Precision matters more than volume.

For organizations pursuing AI-driven automation, understanding those relationships is what transforms information into actionable knowledge.

Unlocking Institutional Knowledge

Another emerging opportunity lies within enterprise archives.

For years, organizations have struggled to preserve institutional knowledge when employees leave or projects conclude. Valuable expertise often remains buried within historical documents, emails, and project records.

Generative AI is changing that equation.

By understanding historical content within its business context, AI can help organizations uncover expertise, preserve institutional knowledge, and make years of accumulated information accessible to future employees and decision-makers.

Rather than functioning as static archives, enterprise repositories become living sources of organizational intelligence.

Governance Remains Essential

As AI becomes more deeply embedded into enterprise workflows, governance becomes increasingly important.

The conversation highlighted that security extends beyond protecting documents. Organizations must ensure AI respects existing permissions, policies, and compliance requirements so sensitive information is never unintentionally exposed through AI-generated responses.

Strong governance ultimately enables organizations to scale AI with greater confidence.

AI Should Work Where Employees Already Work

One of the strongest themes throughout the discussion was that enterprise AI should fit naturally into existing workflows.

Instead of requiring employees to adopt yet another application, AI should surface trusted knowledge inside the platforms they already use, including SAP, Salesforce, Microsoft, Oracle, and other enterprise systems.

When AI operates within familiar business applications and is grounded in trusted enterprise context, organizations can move beyond experimentation toward measurable business value.

Watch the full conversation between Liz Miller and Michael Cybala to learn how OpenText is helping enterprises transform content, context, and knowledge into the foundation for trusted AI.