Topic starter
22/08/2026 12:01 pm
TL;DR: Enterprise AI value is now constrained less by model novelty than by governance, context, and readiness, according to Trust3’s analysis of the Gartner Data & Analytics Summit 2026. The post argues that semantic layers, metadata systems, and policy enforcement must become living infrastructure if organisations want autonomous agents to act on business data safely and usefully.
Editorial analysis by NHI Mgmt Group, based on content published by Trust3: “The Next Wave of Data & AI: Insights from Gartner Data & Analytics Summit 2026”.
By the numbers:
- AI-readiness spending focused on making data usable for GenAI will increase sevenfold between 2025 and 2029, according to Trust3.
Key questions
Q: How should organisations govern AI agents that need database access?
A: Organisations should give AI agents narrow read access by default, separate write privileges from query privileges, and require human approval for any session that can modify data.
Q: Why do AI systems need semantic context layers before production use?
A: Because retrieval alone does not explain business meaning.
Q: When should teams treat AI governance as a runtime control problem rather than a policy issue?
A: As soon as a system can change outputs, access data, or trigger actions after deployment.
Practitioner guidance
- Establish a governed semantic layer for AI inputs Map business definitions, lineage, and classification rules before exposing raw enterprise data to LLMs or agents.
- Extend policy enforcement to agent runtime decisions Require behavioural controls that evaluate what an AI agent can access, when it can act, and which systems it may influence.
- Classify and govern unstructured data before GenAI use Inventory the document, ticket, chat, and file sources that will feed GenAI use cases, then apply classification and access rules consistently across them.
Bottom line: Enterprise AI initiatives now depend on governance, semantic context, and readiness as much as they depend on model capability.
Explore further
View Full Forum → | NHI Foundation Course → | Our Services → | Read the full analysis →
22/08/2026 12:02 pm
Context is becoming the control plane for enterprise AI. The article’s central insight is that raw data alone is not a safe input to GenAI because meaning, lineage, and policy are what make AI outputs governable. That elevates semantic layers and metadata systems from architecture support to security infrastructure. For identity programmes, the parallel is clear: when agents act on business systems, context determines whether access is justified, auditable, and bounded.
A few things that frame the scale:
- Only 44% of organisations have implemented any policies to manage their AI agents, despite 92% agreeing that governing AI agents is critical to enterprise security, according to the 2026 Infrastructure Identity Survey.
A question worth separating out:
Q: What is the difference between access control and data governance in AI environments?
A: Access control decides which identities may reach a system or dataset. Data governance decides how data is classified, monitored, and approved for use. In AI environments, those two disciplines must work together because model input, output, and action paths can expose sensitive data even when the source dataset appears well governed.
👉 Read our full editorial: AI governance and context layers are becoming core infrastructure