Zoom’s AI Companion strategy rests on a simple but hard-to-deliver premise: that a meeting, call, or customer interaction should automatically trigger real work in connected systems, not just produce a summary that someone else has to act on. Delivering that requires reliable orchestration across multiple AI models and enterprise tools. This week, CTO XD Huang publicly named the infrastructure he thinks can do it.
In a LinkedIn post, Huang endorsed OpenAI’s updated Agents SDK, singling out its sub-agent coordination, multi-step workflow management, and handoff capabilities. “That foundation matters for enterprises,” he wrote, “where work spans systems and rarely follows a straight line.”
What OpenAI’s Agents SDK update actually delivers
The SDK update adds two things that have been absent from most enterprise agent deployments: sandboxing and a long-horizon harness.
Sandboxing confines agents to isolated workspaces, specific files and specific tools, nothing else. The new harness is designed for complex multi-step work, providing persistent state and coordination so agents can maintain context across lengthy tasks. OpenAI also confirmed subagent support and code mode are in development for both Python and TypeScript, and the SDK already supports over 100 non-OpenAI models, making it provider-agnostic by design.
These are infrastructure additions rather than end-user features. They are also what separates a proof of concept from something you would run a business process on. The lack of sandboxing has been a genuine blocker for IT teams asked to sign off on autonomous agents running across production systems.
Read more: Agentic AI in enterprise UC
How the SDK fits Zoom’s agentic AI strategy
Zoom has spent the past year building what it describes as a “system of action”: the argument being that meetings and calls should not just produce notes, they should trigger downstream work. At Enterprise Connect in March, it announced custom AI agents capable of orchestrating workflows across Salesforce, Slack, and ServiceNow, alongside expanded agentic capabilities across Zoom Workplace, Zoom Phone, and Zoom CX.
AI Companion 3.0 runs on a federated model, combining Zoom’s own LLMs with third-party models from OpenAI and Anthropic. That multi-model architecture is where Huang’s focus on the SDK’s handoff and sub-agent coordination becomes concrete. A federated system that cannot coordinate cleanly between components does not behave like a coherent product. The SDK’s orchestration primitives address that directly. As Huang wrote:
“The future isn’t AI that responds — it’s AI that moves work forward.”
Where Zoom sits in the wider enterprise agentic AI market
Enterprise Connect 2026 showed the direction of the market. Vendors across UCaaS, CCaaS, and collaboration described autonomous AI agents that interpret user intent, retrieve contextual data, and trigger actions across enterprise systems.
The consistent pressure from enterprise buyers was for measurable results, not pilots. Gartner predicts 40% of enterprise applications will include task-specific AI agents by end of 2026, up from under 5% in 2025. That figure reflects procurement timelines as much as technology readiness.
The SDK’s built-in tracing, guardrails, and observability matter in that context. Governance and auditability are frequently the reason deployments stall at the IT sign-off stage. Building them into the infrastructure layer, rather than leaving them to individual implementations, removes a significant source of friction.
What agentic AI in UC platforms means for enterprise buyers
UC vendors are increasingly building on shared or converging infrastructure. The differentiation will come from how well that infrastructure gets applied where enterprise decisions get made: the call that should update a CRM record, the meeting that should initiate an approval workflow, the support interaction that should close a ticket without anyone carrying it manually between systems.
As Huang put it:
“We’re excited about what this unlocks, from drafting plans to triggering actions across CRM, ticketing, and collaboration tools.”
Analysts covering Zoom’s Perspectives event last week noted that the company’s next phase will be proving agentic workflows deliver consistent ROI across verticals. The SDK makes more of that technically feasible. Whether platforms can execute on it consistently at scale is what the next eighteen months will show.
