The Work is Still Human

Why we need people to stand by the work, not just another platform

A few months back, one of our clients came to us with a problem the board had handed them. A big AI push was on, the kind that turns up in the annual report, and the engineers were expected to deliver it. The catch was that this is a heavily regulated business, where getting it wrong doesn’t result in a bad sprint, it’s a call from the regulator. They had a solid team. They had bought them the platform, the full library, every AI course going. And their people had worked through it, dutifully, badges and all.

Then the real work landed, getting models into production on infrastructure that genuinely mattered, and a lot of it stalled. Not because the engineers weren’t bright. Because the courses had shown them the pieces and nobody had ever shown them how to put the pieces together when it counted, with a regulator looking over their shoulder. The lead who rang me wasn’t cross about it. She’d just worked something out that I think a lot of people are quietly working out right now. She’d bought her team a library, when what they needed was a teacher.

On-demand content is not a capable workforce

This is the story I hear everywhere right now, in some version or another. Companies are racing to build out AI, modernize the infrastructure under it, and secure the whole thing. They’ve bought the platforms, sometimes three or four of them. And then they discover that a full content library and a capable workforce are not the same thing.

The ground is moving. What these teams are being asked to build didn’t exist a year ago, and it’ll change again before they’ve mastered it. Which is exactly why you want a person beside them, not a library behind them.

The work now lives across systems

Part of what makes this moment hard is that the work itself has changed shape. We used to be able to slot a person into a defined role and hand them a course that matched. The work now sits in the seams between systems. The people who move it forward are systems thinkers and operators, the ones who can see how the pieces connect and then actually build the thing and run it when it’s under load. That takes technical depth.

It also takes the parts that don’t fit in a quiz: the judgment to know what to do when the runbook runs out, the confidence to make the call, the instinct that something’s wrong before it breaks. You don’t pick that up from a video at 1.5x speed.

Platforms have a place. Applying the work isn’t it.

To be fair, on-demand options have their place. When you need to refresh a fundamental, prep for a cert, or get a thousand people to a basic baseline, self-paced content is a good tool. Cost-effective, scalable, available at 2am. We’re not here to tell anyone to throw that away.

But there’s a category of work where content on its own leaves you exposed. The work that’s regulated, mission-critical, or simply new, the work that doesn’t have a playbook yet because nobody has written one. For that, people need to learn from people who have already done it. Someone who can watch them apply it, give feedback in the moment, push back when the thinking is off, and stay with them while it sticks. That client’s engineers didn’t need a better course. They needed someone in the room, real or virtual, who had shipped what they were being asked to ship, in a business as exacting as theirs.

In the middle of the biggest automation wave most of us have lived through, the spend on learning from a real expert is going up, not down. IDC’s latest forecast for IT education and training services has the market growing steadily through 2029, with the Americas climbing from roughly $10.3 billion in 2024 to $12.8 billion. And inside that number, instructor-led training delivered virtually keeps gaining ground, with IDC pointing to demand in exactly the areas everyone is scrambling to staff: AI, cloud, and security.

What’s new is already tried and true

When the work gets harder and the stakes climb, organizations reach for people.

This is the whole reason Global Knowledge exists. For thirty years, we’ve believed one thing about how people get good at hard technical work: they learn it from people who have already done it. That belief didn’t change when the cloud arrived, or when security became everyone’s problem, or when AI started rewriting what the work even is. What changed is the rest of the field. The rush to automate learning made it look, for a while, like content could stand in for expertise. It can’t.

So we’re not reinventing ourselves for the AI moment. We’re a global community of around 900+ real experts, people who have built, secured, and operated systems like the ones our clients depend on, and who are authorized by the vendors those systems run on, Microsoft, AWS, Cisco, Palo Alto. We teach in person and virtually, in real lab environments, and we don’t disappear when the class ends. The methods keep advancing. The standard hasn’t moved since the beginning. What’s new is that the moment finally makes it obvious why it matters.

If your AI and infrastructure plans depend on capability your team doesn’t have yet, another platform license won’t close that gap. People will. Real experts who teach your teams to do the work, then stand by whether they can actually do it.

As for that client, we paired their engineers with instructors who had built and shipped the same kind of systems in environments every bit as regulated. A few weeks on, the work that had stalled was moving, models into production, the board’s push back on track. Same people, same ability. They’d just finally learned it from someone who had been there. Build the people. Stand by the work. Everything else is just content.

Darren Bance is General Manager of Global Knowledge.

Learn more about our cybersecurity solutions at: https://www.globalknowledge.com/ca-en/cybersecurity/