The superhero
They know the application, wrote the scripts, remember the architecture and know which service needs restarting when nobody else understands what broke.
AI & Technology
AI changed the how.
Every trade changes when its tools change. AI is another change in the IT toolkit. It does not remove the need to understand the problem — it gives us more time to understand the people behind it.
The Tool Changed
The how has changed before. It will change again.
Carpenters got power tools. Mechanics got computer diagnostics. Engineers moved from drawing boards to CAD. IT moved from physical servers to virtualisation, cloud, infrastructure as code and automation. Now we have AI.
For much of my career, technical value has been closely associated with knowing how to do something: the command, the syntax, the obscure configuration file, or the behaviour of a particular system.
AI changes that dynamic. I can describe what I am trying to achieve, interrogate the problem, test an approach, understand the output and iterate. The skill has not disappeared. The tool has changed.
Context Matters
A technically correct answer is not necessarily the right answer.
AI can give me twenty ways to solve a technical problem. It does not automatically understand which solution is appropriate for the organisation, the customer, the team or the person sitting in front of me.
We still need to understand what we are trying to achieve, where the change belongs, when it should happen, who it affects and why we are doing it.
The how may be easier to find. Judgement and context still matter.
More Human, Not Less
AI can give us time to build relationships.
There is an understandable fear that AI will make technology less human. I think it gives us an opportunity to do the opposite.
If AI reduces the time spent searching documentation, remembering syntax, writing repetitive configuration and producing routine documentation, we can spend more time understanding the person we are trying to help.
What is frustrating you? What would make this easier? What happens if this does not work? What does success look like to you?
Those conversations build something no command-line trick can provide: trust.
People First
Nobody wakes up wanting a Kubernetes cluster.
It is easy in IT to become fascinated by the technology. Servers, networks, containers, monitoring, security, automation and cloud platforms are interesting problems to solve.
But the people we support want outcomes. A finance team does not really want a server. A customer does not really want an API. A colleague does not really want a laptop. They want something that helps them achieve what they need to do.
Our job is to understand that outcome and use technology to help deliver it. AI potentially gives us more time to concentrate on that.
The Superhero Problem
“Call Dave” is not a resilience strategy.
We’ve all worked with them. Maybe I was even one of them.
The hidden risk
Everyone celebrates the person who can always fix it, while the organisation quietly becomes more dependent on one individual's knowledge.
The better model
If one server taking down a service is a design problem, one person's absence preventing operation of a system should be treated as a similar risk.
Shared Knowledge
Knowledge should not be power.
Specialist knowledge can become a form of intellectual property inside a team. Sometimes deliberately, sometimes simply because nobody else has the time to learn it. Knowledge becomes ownership, and ownership can become power.
AI changes the economics of knowledge. It does not magically give everyone twenty years of experience, but access to technical explanations, examples and assistance is becoming dramatically more democratic.
The junior engineer can ask questions they might have been embarrassed to ask. An infrastructure engineer can explore application code. A developer can understand networking. A service manager can better understand what an engineering team is proposing.
The barriers between disciplines become a little lower.
Experience
If answers are easier to produce, judgement becomes more valuable.
Expertise does not become worthless. Experience helps you recognise when something does not look right, understand risk, know which question to ask next and recognise when the technically elegant solution is completely inappropriate for the organisation.
AI can tell you how to make a change. Experience might tell you: do not make that change at 4:30 on a Friday afternoon.
That is a different kind of knowledge, and it remains valuable.
A Level Playing Field
Access to knowledge does not need to be the barrier it once was.
Someone who understands the problem but does not know the exact syntax can now participate. Someone with an idea but without years of programming experience can prototype it. Someone learning a technology can have a concept explained in different ways until it makes sense.
Experienced engineers can combine their judgement with tools that dramatically increase what they can accomplish.
The playing field becomes more level — not because everybody suddenly knows everything, but because access to knowledge is no longer the same barrier it once was.
From Gatekeepers to Guides
Our value does not have to come from being the only person who knows.
Our value can come from helping other people succeed: understanding risk, asking better questions, mentoring, connecting technology to business outcomes, explaining complicated things simply and knowing when not to use the technically clever solution.
Most importantly, it can come from understanding customers and building trust.
That is a healthier form of expertise than protecting a collection of commands and configuration knowledge that nobody else understands.
My Takeaway
We can all be heroes.
AI is changing IT. I do not think the future is one where AI does everything and people become irrelevant. The opportunity is almost the opposite.
Let AI help with syntax, repetitive work, searching, explaining, drafting, comparing and automating. Then use the time it gives us to listen, understand, explain, build relationships, exercise judgement and earn trust.
The tools have changed. They always have. The how has changed, but the what, where, when, who and why still matter — perhaps more than ever.
If knowledge becomes easier to share and technology gives more people the ability to turn ideas into something real, perhaps we do not need to rely on a handful of superheroes anymore.
We can all be heroes.