Red Hat is previewing a set of Model Context Protocol (MCP) servers that it has embedded within multiple platforms while at the same time revealing additional security capabilities that will be included in two forthcoming updates to Red Hat Enterprise Linux (RHEL).

MCP servers have been added to RHEL, the Red Hat Satellite framework for managing RHEL environments and the Red Hat Lightspeed automation framework. That capability will make it simpler for artificial intelligence (AI) agents to invoke these various platforms within the context of a larger natural-language workflow.

Red Hat has also made available additional certified content for RHEL that can be used to reduce downtime and minimize human errors using the Red Hat Ansible automation framework.

At the same time, Red Hat is also making an edition of RHEL that can be deployed using containers starting with versions 10.2 and 9.8 of RHEL along with a technology preview of a sealed images capability to better secure cloud-native application environments.

Finally, Red Hat is adding support for post-quantum cryptography (PQC) for RHEL and Red Hat Certificate System 11.0, which can be used to apply quantum-resistant signatures to workloads, and expanded an alliance with CrowdStrike to provide access to more than 2,300 additional malware signatures to improve the cybersecurity posture of Red Hat environments.

Gunnar Hellekson, vice president and general manager for RHEL at Red Hat, said, in general, there is now an effort underway to balance two often conflicting goals. In some instances, IT teams more than ever need long-term access to stable instances of RHEL that they do not want to update. At the same time, there are use cases where IT teams need access to the latest capabilities. Red Hat is now providing both early access to new technologies while now extending the level of support it provides for operating system releases beyond three years.

In some cases, it’s not unusual for IT organizations to have both requirements for different classes of workloads they may be running, noted Hellekson.

It’s not quite clear yet what impact AI will have on IT operations but at this point fundamental change is all but inevitable. Many IT teams have been relying on machine learning (ML) algorithms for years now to drive AI for IT operations (AIOps). With the rise of agentic AI, it’s becoming possible to now use natural language to assign a wide range of IT tasks to AI agents. There is, of course, still a need for IT administrators to verify those tasks are completed as expected, but in time, many of the manual IT tasks that have historically conspired to make the management of IT tedious are being eliminated.

In the meantime, however, IT environments themselves continue to become more complex. The challenge and the opportunity now is to find a way to strike the right balance between automation and the need for human supervision. After all, the only thing more problematic than a manual task is trying to roll back an automated process without anyone on the IT staff understanding how it was performed in the first place.