Red Hat has taken a significant step toward integrating artificial intelligence into IT automation, releasing its Model Context Protocol (MCP) server for Ansible Automation Platform as generally available and previewing a new automation orchestrator designed to keep AI agents on a short leash. The move aims to give enterprises the ability to harness AI for automating workflows while ensuring that any actions taken are vetted through deterministic, human-approved playbooks.
The MCP server, now generally available, allows any AI tool — including those from Google, Anthropic, OpenAI, and others compatible with the OpenAI API — to connect directly to Ansible. This opens the door for natural language interfaces, where developers or operations staff can request automations in plain English, and the AI will translate those requests into actionable steps. However, Red Hat is emphasizing that this new capability comes with strict guardrails to prevent the kind of AI mishaps that have made headlines in recent months, such as unauthorized database deletions or unplanned production outages.
“AI is unpredictable,” said Sathish Balakrishnan, vice president and general manager of the Ansible business unit at Red Hat. “When you suddenly put AI into your production environment and ask it to change it, you’ve seen the articles about how a company lost its database.” To mitigate that risk, Red Hat is requiring that AI-initiated automations rely on pre-built, tested, and approved playbooks. If the AI proposes something new, a human must be in the loop to verify the action before it is executed.
The new automation orchestrator, currently in technology preview, funnels all AI requests through these standardized playbooks, which are repeatable, deterministic, and significantly less expensive than calling a large language model during runtime. Balakrishnan pointed out that using AI to perform routine tasks like patching a server is wasteful. “We all know tokens are expensive,” he said. “We know the best way to patch a machine — why call an AI to do that when you already have a playbook that’s been in use for ten years?”
This approach addresses a critical concern among IT leaders: the security and reliability of AI agents operating in production environments. Paul Nashawaty, an analyst at Efficiently Connected, described the MCP access as both exciting and risky. “If those agents are connected to highly privileged automation systems, the blast radius can become enormous, including accidental production outages or destructive actions,” he said. He recommended that companies avoid giving AI unrestricted production access or broad admin privileges, and instead focus on use cases like AI-assisted troubleshooting, compliance remediation, developer self-service, and human-approved workflow execution.
Jevin Jensen, an analyst at IDC, noted that he has been waiting for vendors to provide natural-language front ends for their platforms for the past 18 months. “This really broadens the use and value of the platform to new users and improves efficiency of existing users,” he said. However, he stressed the importance of governance, particularly role-based access control, to reduce risk. “It is important — with or without MCP — that enterprises properly utilize and leverage role-based access control,” he said.
In addition to the AI integration, Red Hat announced several other enhancements to Ansible Automation Platform. Administrators can now delegate the ability to trigger automations to end users. For example, factory floor managers can initiate updates at times that cause minimal disruption to manufacturing schedules. Also, multiple events can now trigger the same automation playbook, eliminating the need to create separate playbooks for each event. These features are designed to make automation more accessible and efficient across the enterprise.
The push toward AI-powered automation reflects a broader industry trend. As organizations accumulate more data and complexity, the ability to describe desired outcomes in natural language and have an AI system execute them via validated playbooks could dramatically reduce the time needed for incident response, compliance remediation, and infrastructure scaling. However, the risks are equally real. Without proper oversight, an AI agent with access to a company’s entire automation library could inadvertently cause widespread damage.
Red Hat’s strategy appears to be one of cautious enablement: provide the tools for AI integration, but embed safety mechanisms that ensure human oversight remains central. The company is encouraging customers to start in development environments or low-impact cloud areas before moving to production. This phased approach allows organizations to learn the capabilities and limitations of the AI agents without exposing critical systems to unnecessary risk.
The Ansible Automation Platform has long been a cornerstone of enterprise IT automation, particularly for configuration management, application deployment, and orchestration. By adding MCP support and a controlled orchestrator, Red Hat is positioning itself to meet the growing demand for AI-assisted operations while maintaining the reliability and security that enterprises require. The technology preview of the orchestrator signals that the company is still refining the user experience and guardrails, but the general availability of the MCP server shows that Red Hat is confident enough to let customers start experimenting today.
As AI continues to permeate every layer of IT, the ability to integrate it safely into existing workflows will be a key differentiator for platform vendors. Red Hat’s approach — embracing AI but constraining it through deterministic playbooks and human approval — may well become the blueprint that other enterprise software companies follow. For now, Ansible users have a new set of tools to explore, with the promise of more to come as the orchestrator moves toward general availability.
Source: Network World News