Artificial “Adolescence” Requires a New Mainframe Mindset

Hollywood has long depicted fantastical tech, from Star Trek’s ship computers to Iron Man’s JARVIS. Today, AI feels like one of those cinematic futures brought to life. But unlike the movies, it isn’t a finished product. It’s a fast-moving reality bringing massive productivity gains alongside unprecedented security and operational challenges.

When I recently asked a roomful of technologists if AI had changed their daily work, every single hand shot up. Mine too. I haven’t used Google Search, the greatest consumer “killer app” of the last two and a half decades, in over a year. 

We are all part of an awe-inspiring global shift towards more productivity, higher-value output, and reduced rote work through AI. But this generational shift comes with equally great challenges. The first is a change in mindset, which is always much harder to change than technology. The second involves security, which the mainframe industry is renowned for.

Securing our “adolescent’s” future

Because of AI, the IT industry is grappling with previously unseen security vulnerabilities. Consequently, the industry is rightfully releasing security patches like never before at an unprecedented pace. Although filled with promise and already demonstrating immense value, what we’re dealing with is not “artificial intelligence,” it’s “artificial adolescence.” It’s still developing, growing, and rebelling (if not going rogue) sometimes. Thus, we need tighter security as these powerful agents continue to develop and disrupt the status quo. Inertia and doing what we’ve always done no longer work. 

That generational shift asks a new, soul-searching question of technology leaders: “Do their people have the skills to defend their IT estate and turn AI into powerful profits?” The heartfelt answer will increasingly determine how successfully organizations modernize, staff, and get more value from not only the mainframe—a bedrock platform used by 70% of the world’s leading companies—but their entire business. 

The mainframe’s future, in other words, is as much a workforce story as a technology story. DevX, AI, quantum computing, and modernization matter. But people are what bring those capabilities to market. In fact, people are the only constant in technology. They are the “RI”—real intelligence. We must expand their skills and transfer past experience to conquer tomorrow. 

The job description has changed

For decades, mainframe expertise meant deep knowledge of z/OS, applications, databases, security, and operations. Those skills remain essential. But they are no longer enough.

Developers and operators increasingly must master APIs, automation, modern delivery, AI-assisted development, and security—alongside traditional mainframe disciplines. Developer experience matters because organizations cannot afford to make every new engineer learn the platform through years of institutional knowledge and tribal practices.

The same is true for retention. A mainframe environment becomes vulnerable when critical knowledge exists primarily in the heads of a handful of experienced employees. Retirement, turnover, or even an unexpected absence can become an operational risk.

That need for continued access to knowledge and experience makes your talent strategy a reliability strategy.

Organizations need to document their knowledge, cross-train teams, and give newer employees hands-on experience with real systems. Training matters. So does documenting wisdom and teaching inquisitiveness. But the goal should be broader than completing courses. It should be building a workforce that can confidently operate, develop, and modernize the mainframe.

AI as a mainframe skill

AI is also moving from something interesting to something practical.

It is not replacing the mainframe. Instead, it is becoming another way to develop, troubleshoot, automate, and extract more value. AI-assisted tools can help developers understand unfamiliar code, identify problems, and accelerate routine work. For operations teams, AI can help surface patterns and make large environments easier to manage.

That changes what organizations should expect from mainframe professionals. Waiting for a narrowly defined “AI for mainframers” role misses the larger opportunity. The more valuable skill is knowing how to apply AI to existing mainframe expertise.

The same principle applies to automation. If a person repeatedly provisions an environment, performs a deployment step, checks a routine condition, or carries out a maintenance task, the natural question becomes, “Why is a person doing this?”

Automation is increasingly table stakes, not a nice-to-have. By reducing repetitive work, automation allows skilled employees to solve more complex problems, improve systems, and produce more profitable outcomes.

The talent multiplier

That’s where technology becomes a talent multiplier.

Modern developer experiences can make the mainframe more accessible. Automation can amplify the output of teams. AI can help employees work faster and cut through complexities. And new interfaces can connect mainframe capabilities to the broader business.

But none of those technologies creates business value on its own. People do. The mainframers best positioned for the future will combine deep z/OS knowledge with modern AI, automation, and security. And they will understand the business context behind the systems they build. 

That is the real evolution of the mainframe workforce. Moving from “I know mainframe” to “I know how to leverage mainframe for the benefit of the entire business.”

The good news is new mainframers are joining our ranks like never before to meet this unprecedented challenge. They want to work in this field. They understand that in technology not everything new is good and not everything “old” is bad. That’s a big reason why recent investment in mainframes has been remarkable.

Looking forward

For technology leaders, the imperative is clear. Of course, invest in the platform. But invest even more in the people who carry it forward. Build skills before they become gaps. Automate before manual work becomes a bottleneck. And make knowledge transferable before it walks out the door.

The mainframe is evolving, probably faster than ever before. The organizations that get the most from it will be those that make their people part of the transformation—not an afterthought or casualty to it.