By Senior Contributor,Tony Bradley

Copyright forbes

AI workloads are set to surge 50% in the next two years, but many cloud teams admit they’re already stretched thin and struggling to keep up.

The AI boom isn’t some future wave waiting offshore—it’s already breaking over enterprise infrastructure, and most cloud teams are barely treading water.

That’s the blunt picture painted by ControlMonkey’s recent GenAI Readiness Report. The survey of 300 senior leaders in DevOps, cloud engineering and infrastructure roles found workloads driven by generative AI are set to surge 50% in the next 12–24 months. Nearly four in ten leaders said the increase would be “significant” or even “exponential.”

AI isn’t just another category of workload—it’s quickly becoming the defining workload for enterprises. But while the demand curve is steep, the readiness curve is flat. That mismatch is creating pressure on already-stretched cloud teams, who are being asked to support innovation at a pace faster than their systems—and often their skills—can sustain.

Bandwidth Is the First Casualty

The most striking finding from ControlMonkey’s research is that 46% of DevOps and cloud leaders say their teams simply don’t have the bandwidth to innovate. They’re already consumed by keeping the lights on.

That’s a strategic threat, not just a technical one. Enterprises can’t afford to spend all their cycles firefighting today’s outages and configuration drift while leaving no room for tomorrow’s innovation. In the face of an AI surge, teams with no innovation capacity risk falling behind not just competitors, but the expectations of their own business units.

From my own conversations with CIOs and CISOs, this pressure is intense. Most leaders want to push AI projects forward, but too often their teams are so busy plugging gaps that innovation takes a back seat.

MORE FOR YOU

Automation Isn’t Keeping Up

When asked about automation readiness, only 46% of teams said they were fully prepared for AI-driven workloads. The rest admitted they weren’t ready for scale.

Automation is supposed to be the lever that lets small teams support big ambitions. But familiar gaps—reliability (43%), skills shortages (39%) and scalability limits (36%)—are still blocking progress. These aren’t exotic “AI problems.” They’re old, boring infrastructure problems that haven’t been solved, and now they’re colliding with AI’s breakneck pace.

It’s still relatively early when it comes to AI, but things are moving fast and those who are not prepared risk getting left quickly in the dust. GenAI doesn’t change the fundamentals of cloud operations. It just magnifies them. If your automation pipelines are fragile now, they’ll break faster under AI-driven demand. If your monitoring is patchy today, blind spots will multiply tomorrow.

Visibility, Costs and Governance

The report also highlights the usual pain points that have plagued cloud adoption for years: costs, visibility and governance. Thirty-seven percent of leaders cited rising costs as their top infrastructure barrier. Another 36% pointed to lack of real-time visibility and 32% admitted they struggle with effective resource allocation.

Again, these issues aren’t new. But the context is. In a pre-AI world, inefficiency meant budget overruns or slower releases. In an AI-driven world, inefficiency becomes existential. When workloads spike unpredictably, the inability to see what’s happening—or to trust that automation will respond—can grind innovation to a halt.

Governance is another looming choke point. Nearly a third of respondents named security governance and compliance complexity as top challenges. AI workloads often touch sensitive data and introduce new dependencies, making governance a moving target. Without standardized policies, enterprises risk either slowing innovation with red tape or accelerating recklessly into compliance disasters.

Why Skills and Visibility Matter Most

Perhaps the most revealing data point is what leaders say would actually help. Training and visibility top the list, at 45%. Cost control (21%), governance (20%) and automation (14%) rank lower.

That’s telling. Enterprises don’t necessarily want more tools—they want more clarity and expertise. You can’t automate what you don’t understand, and you can’t govern what you can’t see. I’ve seen this pattern play out before: teams buy platforms with promises of speed and scale, only to realize they lack the people and visibility to make those tools effective. The technology itself isn’t the bottleneck—it’s the human element.

The survey also revealed that ownership of AI tooling is fragmented. In most organizations, responsibility is spread across three or more groups—engineering, DevOps, C-suite and data science. In 12% of enterprises, five or more functions share ownership. That kind of diffusion slows decisions and creates misalignment. Centralizing ownership—ideally with the teams closest to the infrastructure—may be one of the simplest steps organizations can take to speed readiness.

Closing the Cloud Skills Gap

The survey’s emphasis on training and expertise underscores a growing reality: success in scaling AI depends as much on people as it does on platforms.

That context helps explain why ControlMonkey recently introduced KoMo AI, a GenAI-based tool aimed at easing Infrastructure-as-Code bottlenecks. Rather than replacing engineers, KoMo is designed to generate Terraform code aligned with an organization’s existing modules and policies, with the goal of reducing repetitive reviews and helping less-experienced team members contribute more effectively.

At its launch, CEO Aharon Twizer described KoMo as a way to “close the cloud skills gap by evolving self-service.” Whether KoMo delivers on that promise remains to be seen, but the timing reflects the urgency highlighted in ControlMonkey’s own research: nearly half of DevOps leaders say they lack the bandwidth to innovate just as AI workloads are set to double.

The Next 12 Months Will Decide

Richard Stiennon, chief research analyst at IT-Harvest, put it in blunt terms: “AI adoption is as important today as getting on the Internet was 30 years ago. Don’t let budget or lack of people get in the way. Empower the curious to surface results quickly.”

That’s the reality facing enterprise IT. GenAI workloads are scaling whether teams are ready or not. The question is whether organizations will invest in the fundamentals—automation that actually scales, visibility that reveals the real picture and skills training that frees engineers to innovate.

Every technological shift eventually stress-tests infrastructure. Virtualization did it. Cloud did it. Now AI is doing it faster and harder. The difference this time is the velocity. Enterprises don’t have years to adapt. They have months.

The companies that survive this inflection point won’t necessarily be the ones with the most advanced AI models. They’ll be the ones whose infrastructure teams can absorb the surge without breaking. In other words, success in the age of AI may depend less on AI itself, and more on the readiness of the people and systems asked to support it.

Editorial StandardsReprints & Permissions