OpenAI has released Secure MCP Tunnel, an enterprise feature enabling supported products—ChatGPT web, Codex, Responses API, and AgentKit—to connect to private or on-premises MCP servers via a customer-hosted tunnel client without exposing those servers to the public internet. The feature addresses a key enterprise adoption barrier by allowing MCP-based tool integrations to remain within private network boundaries. This is a meaningful infrastructure addition for organizations deploying AI agents against internal systems.
OpenAI launched ChatGPT Enterprise, a tier targeting business customers with enterprise-grade security, privacy guarantees, and access to the most capable ChatGPT model available at the time. The product is positioned as a managed deployment option for organizations requiring data protection and compliance features. This marks OpenAI's formal entry into the enterprise SaaS segment with a dedicated product tier.
Anthropic has released the Model Context Protocol (MCP), an open standard enabling secure, two-way connections between AI assistants and external data sources such as business tools, content repositories, and development environments. The protocol introduces a client-server architecture with SDKs, local MCP server support in Claude Desktop, and a repository of pre-built connectors for systems like GitHub, Slack, Google Drive, and Postgres. Early adopters include Block and Apollo, with development tool companies Zed, Replit, Codeium, and Sourcegraph integrating MCP into their platforms. The goal is to replace fragmented, per-source integrations with a single universal protocol, improving context availability for AI agents.
OpenAI is rolling out Workspace Agents to ChatGPT Business and Enterprise customers, enabling organizations to build, schedule, and share agents for repeatable business workflows. Agents can connect to tools including Google Drive, Google Calendar, Slack, and SharePoint, and support custom MCP servers. The feature ships off-by-default with admin controls, and is excluded from EKM-enabled Enterprise workspaces at launch. This represents OpenAI's direct push into enterprise workflow automation, competing with established RPA and no-code automation platforms.
OpenAI is rolling out Workspace Agents to ChatGPT Business and Enterprise customers, enabling organizations to build, schedule, and share agents for repeatable business workflows. Agents can connect to external apps including Google Drive, Google Calendar, Slack, and SharePoint, and support custom MCP servers. The feature includes template-based creation, version history, analytics, and admin controls, with Slack integration for running agents in connected channels.
OpenAI has released 'Company Knowledge,' a feature that integrates context from enterprise applications directly into ChatGPT to provide business-specific answers with citations. The feature includes security, privacy, and admin controls, and is available to ChatGPT Business, Enterprise, and Edu users. This represents OpenAI's continued push into enterprise RAG-style workflows, connecting organizational data sources to its flagship chat product.
OpenAI announced expanded enterprise capabilities for API customers, including enhanced security features and controls, updates to the Assistants API, and new cost management tools. The announcement targets enterprise adoption by addressing common organizational requirements around security, compliance, and budget oversight. No specific model capability changes are described.
OpenAI has extended ChatGPT Workspace Agents to eligible Enterprise workspaces using Enterprise Key Management (EKM). The feature allows organizations to build, schedule, and share agents for repeatable business workflows across connected apps including Slack, with support for custom MCP servers, skills, files, and version history. Agents are off by default and require admin enablement, signaling a cautious enterprise rollout posture.
IBM has open-sourced mcp-context-forge, a Python-based AI gateway, registry, and proxy that sits in front of MCP, A2A, or REST/gRPC APIs and exposes a unified endpoint with centralized discovery, guardrails, and management. The tool is designed to optimize agent and tool calling workflows and supports plugins. With ~3,800 GitHub stars, it represents a notable infrastructure contribution to the MCP/A2A ecosystem from a major enterprise vendor.