Isometric illustration of the IBM watsonx orchestrate Hub UI

IBM watsonx Orchestrate advances enterprise AI agent management with new capabilities at TechXchange 2026 

Discover and manage agents across platforms, continuously improve performance and establish the security and controls needed to scale agentic AI across the enterprise.

AI agents are rapidly becoming part of how enterprises operate. Organizations are building agents to automate workflows, improve employee productivity and transform customer experiences. But as adoption accelerates, a new challenge is emerging: managing the growing number of agents developed and deployed across different teams, tools, frameworks and cloud environments.

Building an AI agent is only the beginning. Enterprises also need to know which agents exist, understand how they perform, control what they can access, govern what they can do and continuously improve their reliability. Without a consistent approach to managing agents throughout their lifecycle, organizations risk replacing isolated AI experiments with a fragmented and increasingly complex agent ecosystem.

As a result, agent management platforms (AMPs) are becoming essential to enterprise AI strategies.

IBM watsonx Orchestrate® is designed to address this challenge as an open, hybrid and trusted enterprise agent management platform. It brings together the capabilities organizations need to orchestrate, manage, govern, secure and optimize AI agents throughout their lifecycle, regardless of where those agents were built or run. 

At IBM TechXchange™ 2026, we’re advancing this vision with new capabilities that help organizations discover and manage agents across platforms. These capabilities also enable teams to continuously evaluate and optimizeagent performance while establishing stronger security, identity and policy controls.

Together, these innovations extend the watsonx Orchestrate agentic control plane and help enterprises move from fragmented agent deployments toward a more connected, governed and reliable approach to agentic AI.

Manage AI agents across your entire enterprise—not just one platform

Enterprise AI doesn’t happen in a single ecosystem. Different teams choose different technologies based on their business needs, existing investments and preferred development tools. As a result, organizations often operateagents built with IBM, Amazon Bedrock, Microsoft, Google and custom frameworks across their environments.

The challenge is that most platforms provide visibility and management primarily within their own ecosystems. For enterprises adopting multiple technologies, this platform-centric approach can create fragmented inventories, inconsistent controls and limited visibility into the broader agent estate.

With watsonx Orchestrate, enterprises can manage agents built in different environments through a single experience without rebuilding them or standardizing on one platform.

At TechXchange, we’re introducing the watsonx Orchestrate AI Gateway, a single enforcement layer for discovering third-party agents, models and Model Context Protocol (MCP) servers across platforms such as Amazon Bedrock, Microsoft and Google. It governs those assets through consistent observability, evaluation and optimization. Customers can now interact with those third-party agents directly through watsonx Orchestrate chat and channels. 

In addition, the watsonx Orchestrate catalog is evolving into a governed catalog that helps teams discover, share and reuse approved agents and tools rather than repeatedly building the same capabilities. The governed catalog surfaces evaluation data at the moment of discovery and is aligned to OWASP, CoSAI and the IBM AI Risk Atlas. For select agents such as Ziemacs, Cognos Partners and others, builders and admins can see performance, reliability and latency metrics. They can also view risk signals, relevant controls and compliance information such as the EU AI Act.

For select MCP servers such as Box, Visier, New Relic and others, the catalog surfaces MCP protocol conformance, tool schema integrity, tool description safety, input validation and default credential rejection. These capabilities transform the catalog into a living quality assurance system for finding, trusting and leveraging the right agentic AI assets.

We’re also strengthening the ability to manage the broader ecosystem of resources agents depend on, including models, tools and MCP servers.

Today’s enterprise agents do not simply generate text. They run custom code, invoke third-party tools through the MCP, coordinate multi-agent graphs and evaluate real-time transactions against corporate backends. When AI agents execute dynamic tool code and integrate with external community MCP servers, shared-kernel infrastructure presents significant enterprise risks. 

The watsonx Orchestrate Premium Runtime provides a dedicated, multi-backend hosting, isolation and metering layer. It addresses enterprise risk through hardware-virtualized container isolation, zero lateral movement and hardware-bound resource containment.

For enterprises, this capability means greater visibility into existing AI investments, less duplication and a more consistent approach to managing agents as adoption expands. By providing choice without fragmentation, organizations can continue building with the technologies that work best for them while establishing centralized visibility, management and control across their agent ecosystem.

Move beyond monitoring to continuously improve agent quality and performance

Knowing which agents exist is important. Knowing whether those agents are performing reliably is equally important.

As organizations move agents into production, traditional approaches to testing and monitoring are no longer enough. AI agents execute tasks, invoke tools and interact with enterprise systems in ways that can be difficult to predict.

Enterprises need a way to evaluate agent quality before deployment, understand agent behavior in production and use those insights to continuously improve performance.

watsonx Orchestrate brings evaluation, observability and optimization together across the agent lifecycle—helping organizations move beyond simply identifying problems to improving agent outcomes. 

New capabilities include:

  • Custom LLM-as-a-Judge evaluation: Define evaluation criteria in natural language based on what matters to the business, rather than relying solely on generic quality metrics.

  • User simulation and deterministic evaluation: Define and simulate test journeys with grounded success criteria, verifying tool calls and parameters to identify potential failures before production.

  • Trace Inspector: Gain detailed visibility into agent execution, including the steps, actions and tool interactions involved in completing a task. 

  • AgentOps Agent: Investigate agent failures, generate test cases and automatically optimize the agent’s quality and performance before promotion through a conversational experience.

Together, these capabilities establish a continuous improvement cycle: evaluate, observe, diagnose, optimize and verify. Instead of treating evaluation as a one-time development activity or observability as a passive dashboard, organizations can use insights from agent behavior to drive ongoing improvements in quality, reliability and efficiency.

This framework is an important step toward operating AI agents with the same rigor and accountability enterprises expect from business-critical systems.

Build trust into every agent with enterprise-grade security, identity and controls

As AI agents take on more responsibility, they increasingly interact with sensitive information, invoke enterprise APIs and perform actions across business systems.

That creates a fundamental requirement: organizations must be able to establish who or what an agent is, what it is authorized to do and how its behavior is governed.

Without consistent identity and security controls, agents can become overprivileged, access resources beyond their intended scope or operate without sufficient accountability.

watsonx Orchestrate helps enterprises embed security and governance into agent operations, applying consistent controls across agents, models and tools. 

At TechXchange, we’re advancing these capabilities through expanded policy enforcement, configurable controls and integration with enterprise identity and security systems.

Organizations can define and enforce controls for agent behavior, including content guardrails and usage restrictions, while using AI Gateway capabilities to extend supported policies across connected agent environments.

These capabilities help organizations establish consistent operational boundaries without requiring every agent to originate from the same development platform.

We’re also advancing agent identity and access management as part of IBM’s broader approach to securing enterprise agentic AI. 

As announced at TechXchange, IBM Agent Identity and HashiCorp® Vault® complement watsonx Orchestrate to help enterprises establish governed agent identities and manage access to business systems. IBM Agent Identity provides identity and authorization context, while HashiCorp Vault supports secure, least-privilege credential access. Together with watsonx Orchestrate, these technologies support an approach in which organizations can identify agents, establish appropriate access, apply operational controls and maintain accountability as agents perform work across enterprise systems.

IBM watsonx Orchestrate can also connect directly with Microsoft Entra and IBM Verify as identity providers to manage both human and non-human agent identities.

The objective is not to restrict what agents can do, but to establish the trust and security foundation needed for agents to take on greater responsibility safely.

One platform for the complete AI agent lifecycle

The latest innovations in watsonx Orchestrate reflect a broader shift in enterprise AI. The challenge is no longer simply building an agent quickly. It is managing hundreds or thousands of agents across the business, ensuringthey behave reliably and securely and continuously improving the value they deliver.

Addressing that challenge requires a platform that connects the entire agent lifecycle. With watsonx Orchestrate, enterprises can build and reuse agents while managing others created across different environments. They canevaluate and monitor performance, optimize behavior and apply the governance and security controls required to operate agents in production.

This approach is grounded in three principles:

Open: Manage agents built with IBM, third-party, custom and open source technologies without being locked into one vendor ecosystem. 

Hybrid: Support enterprise AI operations across cloud, on-premises and hybrid environments while protecting existing technology investments.

Trusted: Embed evaluation, security, identity, policy enforcement and governance into how agents are managed throughout their lifecycle.

Your agents run everywhere. Manage them all with IBM watsonx Orchestrate.

The innovations announced at TechXchange 2026 represent another step toward making enterprise-scale agentic AI more manageable, reliable and secure, helping organizations move beyond experimentation and put AI agents to work with confidence.

Learn more about IBM watsonx Orchestrate

Author

Suzanne Livingston

Vice President

IBM watsonx Orchestrate