SAN FRANCISCO – Across all sectors,
AI has transitioned from experimentation to mass adoption, and it’s no different in the IT sector.
This has been impossible to ignore at the
RSA Conference (RSAC) this year, where keynotes and breakouts have centered on the operational realities of autonomous systems and the pressure on already strained IT teams. The question on IT leaders’ minds has moved from "How do we monitor every alert?" to "What if our infrastructure could solve its own problems?”
[
SC Media Perspectives columns are written by a trusted community of SC Media cybersecurity subject matter experts. Read more Perspectives here.]
IT’s existing model of operations has historically relied on reactive measures, meaning teams need to stay on-call around-the-clock in case networks are disrupted or compromised, all in the efforts to prevent any downtime. IT has been suffering an operations crisis as a result thanks to the tool sprawl, talent shortages and burnout it’s facing.
In fact,
61% of IT professionals are changing their network configurations on a weekly basis and
60% are feeling at least moderately burned out.
So, how can IT leaders harness AI to implement fully autonomous IT infrastructure that can reconfigure, troubleshoot, and patch networks without oversight? And, what infrastructure do we need to put in place first for organizations to get there successfully?
The burnout crisis
In 2026, IT teams find themselves in a reactive downward spiral, constantly firefighting and handling incidents. This leaves little time to build the strategic initiatives needed to power real AI transformation beyond basic chatbot functionalities.
Tool sprawl compounds the growing burnout problem facing IT teams across all sectors. Organizations are continuing to add monitoring and management tools expecting efficiency gains and workload relief, but end up creating another full-time job for IT pros altogether. Our research found that nearly
50% of IT professionals report having to manage 10 or more network tools.
As a direct result, the IT sector has been struggling.
Most IT employers worldwide (76%) say they struggle to find the tech talent they need, according to a recent report by ManpowerGroup. Workforce challenges including burnout, an aging IT workforce, and talent shortages mean fewer employees are expected to manage increasingly complex environments. IT teams are asked to do more with less, always an unsustainable approach.
IT automation mostly operates in a "see and tell" mode today, where systems detect issues and alert humans to take action. However, systems haven’t historically been able to close the loop independently, and fix the problem themselves.
Autonomous remediation represents the critical shift to the "do" phase with agentic systems that detect, diagnose, remediate, and verify without requiring human intervention for routine issues. In practice, this means infrastructure could handle routine troubleshooting, perform scheduled maintenance, and complete basic tasks automatically. This isn't about replacing human judgment for critical decisions, but about freeing humans from the repetitive, time-consuming tasks that don't require strategic thinking.
An honest assessment
IT has heard automation promises before, and many have fallen short, so a healthy skepticism is understandable. However, there’s been a fundamental shift. We’re seeing a convergence of mature AI and agentic capabilities that understand context and make decisions, with unified visibility platforms that provide comprehensive infrastructure insight, and proven automation frameworks.
IT pros now leverage AI to gain comprehensive visibility across the entire IT environment, but we're also seeing the integration layer that connects these disparate capabilities into cohesive autonomous systems rather than just another collection of single solutions, and that's what agents deliver.
We can now leverage agents to recognize patterns, make decisions and run automated platforms that are sophisticated enough to remediate at scale. For instance, if a database runs slow, an agent can diagnose the root cause and take corrective action automatically.
Conversations happening across RSAC this week reinforce how quickly autonomous capabilities have gone from a distant aspiration to reality. During Monday’s keynote, “
Ambient and Autonomous Security: Building Trust in the Agentic AI Era,” Microsoft Security's Vasu Jakkal said autonomous, self-healing systems are no longer theoretical.
Prepare for the shift
Discussions on the RSAC floor have centered around the reality that the tech has arrived, but how ready are organizations for autonomous systems? Teams will need to establish unified network and data visibility across the IT environment because autonomous systems cannot manage what they cannot see.
IT teams can begin by identifying candidates for autonomous remediation: repetitive, well-understood tasks where automation delivers clear value without significant risk. Security teams must also play a role in establishing governance frameworks with clear guardrails and oversight before deploying autonomous capabilities widely.
Employee preparation matters as much as technical and data readiness. Teams must understand how autonomous systems will change work and prepare to work alongside these capabilities.
AI will become the new language of IT, and employees should have a command over prompting and logic engineering to help agents understand important workflows, as well as metadata tagging and data management to ensure agents are working with the right information.
That said, even the best teams only succeed if the underlying environment gives autonomous systems the context they need to act safely, and that’s why visibility becomes the other half of the readiness equation.
Wednesday’s session “
Attack Surface Everywhere—All Defenders Need Multi-Layer Signals to Keep Up,” from Broadcom’s Jason Rolleston and Eric Chien highlighted how smaller teams can only keep pace by unifying signals across endpoints, identities, and data repositories. That same visibility represents the foundation autonomous systems need to operate safely, make informed decisions, and avoid cascading failures.
With that base in mind, the pace of AI advancement raises the stakes even further. AI models are developing faster than we can keep up, making autonomous infrastructure increasingly within reach.
It's now about how organizations will prepare for this shift, with the right data management infrastructure in place. In the future, IT teams will use AI to build infrastructure intelligent enough to manage itself. This promises to free us up to focus more on strategy, innovation, and driving business value.
Doug Murray, chief executive officer, AuvikSC Media Perspectives columns are written by a trusted community of SC Media cybersecurity subject matter experts. Each contribution has a goal of bringing a unique voice to important cybersecurity topics. Content strives to be of the highest quality, objective and non-commercial.
