ModelOp has released ModelOp Center 3.5, an update to its enterprise AI governance platform that adds deeper integrations with Microsoft and Amazon AI environments. The release connects agents built with Microsoft Copilot Studio and Foundry, Microsoft Agent 365, Amazon Bedrock AgentCore and Amazon Quick to a centralized system for AI discovery, lifecycle management and oversight. The update also expands usage tracking by business use case, giving enterprises a way to connect AI consumption with the applications and models supporting specific business objectives.
ModelOp Center 3.5 Targets Governance Across AI Platforms
ModelOp has introduced ModelOp Center 3.5, an update designed to help enterprises discover, manage and govern AI agents deployed across multiple development and execution platforms.
The release adds integrations across Microsoft’s and Amazon’s AI ecosystems, including Microsoft Agent 365, Copilot Studio and Foundry, alongside Amazon Bedrock AgentCore and Amazon Quick. ModelOp says the integrations bring agents into a centralized system of record, where organisations can connect AI assets to business use cases, assign ownership, apply controls and manage lifecycle workflows.
The launch addresses a growing enterprise challenge: AI systems are increasingly developed across different teams, vendors and environments, making it harder to maintain a consistent view of what is running, why it exists and whether it meets internal policies.
A company might have business teams creating agents with low-code tools, developers building more sophisticated agents and AI teams evaluating models from several providers. Each environment may provide its own management features, but enterprises still need a cross-platform view of their AI portfolio.
ModelOp’s strategy is to provide that oversight layer without requiring businesses to standardise on a single AI execution platform.
From AI Discovery to Lifecycle Governance
ModelOp Center 3.5 extends the company’s AI discovery service across the newly supported environments. According to ModelOp, the service compares discovered agents and other AI assets with an enterprise’s approved inventory to identify unregistered systems, including potential shadow AI.
Discovery is only the first step. The platform is designed to initiate governance workflows when a new agent is found. These can include registering the asset, assigning an initial risk tier, mapping controls based on that risk and routing approvals to the appropriate governance or control teams.
The approach aims to connect technical inventory with operational accountability. Instead of simply recording that an agent exists, organisations can associate it with a business purpose, owner, required controls and lifecycle stage.
That information can help governance teams distinguish between AI experiments, approved internal tools and systems supporting important business processes. It may also make it easier to identify assets that have been deployed without completing the organisation’s review requirements.
However, the effectiveness of automated governance depends on the completeness of discovery, the quality of asset metadata and the accuracy of the policies configured by each enterprise. The announcement does not specify the full set of supported discovery methods or provide independent measurements of how reliably the platform detects unregistered agents.
Connecting Governance Across Microsoft and Amazon
The new integrations target a heterogeneous AI environment in which organisations may use multiple vendors for agent development, model access and execution.
Microsoft Agent 365, Copilot Studio and Foundry support different parts of Microsoft’s enterprise AI environment, while Amazon Bedrock AgentCore and Amazon Quick serve different agent-development and business-use contexts within Amazon’s ecosystem.
ModelOp positions its platform as a layer above these environments. Its system is intended to manage AI assets regardless of whether they were created through low-code, no-code or professional development workflows.
The distinction is important for enterprises that want business teams to experiment with AI without losing central oversight. A shared governance platform could help IT, risk and business leaders work from the same inventory while allowing engineering teams to retain their preferred tools.
ModelOp also says the platform can track frontier models, including those from Amazon Bedrock and OpenAI, as well as MCP tools hosted across multiple servers. The company describes its system as supporting machine learning, generative AI, agentic AI and vendor-provided AI assets.
The announcement does not provide a detailed compatibility matrix, integration limitations or technical documentation showing which metadata and control actions are available for every connected platform. Those details will matter when enterprises assess how much governance can be automated versus handled through manual processes.
Business Use Cases Become the Unit of Governance
One of the update’s central features is the connection between AI assets and the business use cases they serve.
An organisation may deploy several agents and models to support a single process, or use the same model across different business functions. Tracking these assets only by platform or model name can make it difficult to determine who benefits from them, who is accountable for their operation and where costs accumulate.
ModelOp Center 3.5 is designed to associate agents and models with a use case, its owner, applicable controls and lifecycle stage. That approach treats AI as a portfolio of business activities rather than a collection of disconnected technical components.
For governance teams, the model could improve accountability by linking controls to the purpose for which an AI system is used. For business leaders, it could provide a clearer view of which initiatives have been approved and which require further review.
The platform’s usefulness will depend on organisations maintaining accurate use-case records and ownership assignments as agents evolve. The announcement does not describe how the system handles duplicated use cases, agents shared across departments or significant changes to an agent’s purpose after approval.
AI FinOps Shifts Attention to Business Consumption
ModelOp Center 3.5 also expands AI usage tracking and reporting. ModelOp says token consumption can be attributed to business use cases and models, allowing leaders to compare usage across providers and receive alerts when consumption changes materially.
This connects AI FinOps—the monitoring and management of AI-related spending—to business operations. Token-level reporting can show how much a model is used, but assigning consumption to a business use case may help explain why the usage exists and which activities account for it.
That distinction becomes increasingly relevant when agents operate repeatedly through automated workflows rather than being used only in individual conversations. A workflow that runs frequently or invokes several models may generate consumption that is difficult to assess from a single provider’s dashboard.
Use-case attribution could support cost reviews and model comparisons, although token consumption alone does not establish business value. Enterprises still need to consider model quality, infrastructure charges, operational overhead and the outcomes delivered by each use case.
ModelOp has not disclosed quantified cost savings or comparative financial results from the new capabilities.
Governance Platforms Face a Cross-Vendor Test
Enterprise AI governance is developing alongside agent-building platforms, IT service management tools, governance-risk-and-compliance systems and cloud-native control services. Each can cover part of the management problem, but enterprises with diverse AI estates may need a consolidated view spanning multiple providers.
ModelOp says it works alongside existing IT service management, security, data and GRC systems. It also cites a high ranking for the AI Agent Governance use case in Gartner’s 2026 Critical Capabilities for AI Governance Platforms report. That ranking is the company’s stated positioning and should not be interpreted as a Gartner endorsement.
The broader challenge remains practical: a governance platform must discover assets reliably, keep ownership and risk information current, apply appropriate controls and provide evidence that policies are being followed.
ModelOp Center 3.5 expands the company’s cross-platform approach, particularly across Microsoft and Amazon environments. Whether it can deliver consistent oversight at enterprise scale will depend on integration depth, implementation quality and how well its governance workflows fit each customer’s operating model.
Market Landscape
Enterprise AI governance is shifting from model approval toward oversight of agents, tools, workflows and the business processes they support. As organisations adopt AI across multiple cloud providers and development environments, maintaining a complete inventory and consistent controls becomes more complex.
Three trends are shaping the market:
- Cross-platform discovery: Enterprises need to identify approved and unregistered AI assets across different vendors, teams and development environments.
- Lifecycle-based governance: Asset registration, risk assessment, ownership, approval and ongoing monitoring are becoming connected parts of enterprise AI management.
- AI FinOps: Linking model and token usage to business use cases can help organisations understand where consumption occurs, although usage alone does not prove business value.
ModelOp’s strategy is to provide a central governance layer across existing AI platforms rather than replace them. Its differentiation will depend on the breadth and depth of its integrations, the reliability of discovery and the extent to which governance workflows can be automated.
Top Insights
- Microsoft and Amazon integrations: ModelOp Center 3.5 connects AI agents across several Microsoft and Amazon environments to a centralized governance system.
- Automated lifecycle workflows: Newly discovered agents can enter processes for registration, risk tiering, control mapping and approval routing.
- Business context: Agents and models are linked to use cases, owners, controls and lifecycle stages to support portfolio-level oversight.
- AI FinOps capabilities: Usage tracking attributes token consumption to models and business use cases, with reporting and alerts for material changes.
- Governance effectiveness remains a test: The announcement does not provide independent discovery-accuracy results, integration benchmarks or quantified cost savings.
Power Tomorrow’s Intelligence — Build It with TechEdgeAI
