With all the hype and alarmism surrounding AI’s impact on the jobs market, it’s no surprise that workers are looking for ways to protect their roles. In fact, recent research suggests that as many as 35% of employees are now ‘gatekeeping’ their skills for fear of being replaced, with 38% reluctant to train colleagues in the hope it will help safeguard their positions.
On the one hand, it’s an understandable, natural reaction. When technology and automation feel like a threat, the instinct is to tighten control over what you know and make yourself indispensable. But in reality, this approach is fundamentally flawed – and in the context of AI, it may even accelerate the very outcome people are trying to avoid.
Job security in this new AI age won’t come from what you know; it will come from how effectively you can orchestrate what the technology does. And that distinction – between what you know and what you can enable – is becoming increasingly important as time goes on.
From task executor to workflow orchestrator
From my experience, the heart of the issue is a misunderstanding about where human value now lies.
The fear of being replaced by AI is only justified if your value is tied strictly to repeatable, manual tasks. Gatekeeping these skills only creates a false sense of security and further tethers you to the very ‘grunt work’ AI is designed to consume. In that sense, it makes the outcome of automation more likely.
In other words, if your role is built around processes that can be standardised, replicated, and scaled, then those are precisely the workflows AI will target first. Protecting them doesn’t future-proof your career, but it does anchor you to a shrinking part of the value chain.
The real shift people need to make is moving from a ‘task executor’ to a ‘workflow orchestrator’. By handing off the redundant, boring parts of a role to AI, employees free themselves for high-level strategic thinking – the nuanced, creative work that is difficult to automate and far more rewarding.
This is where the opportunity lies. AI can handle the predictable; humans can focus on the exceptional. And crucially, it’s about working with the technology, not trying to compete against it.
The hidden risks of knowledge hoarding for employers
While the individual motivations behind gatekeeping are clear, the organisational consequences are often overlooked – and they can be significant.
For employers, skill hoarding can lead to a number of risks. For one, it creates knowledge silos and single points of failure, where a team is paralysed if the expert who has that knowledge leaves. It also risks making organisations less transparent, where people don’t check each other’s work, which can lead to inefficiency and ‘invisible’ errors. Finally, it can also lead to ‘human bottlenecks’ where experts refuse to train the systems meant to assist them, limiting the capacity for growth.
What’s more, these are all challenges that can worsen and compound over time. A single bottleneck might slow a process, but if there are multiple bottlenecks, it can stall an entire organisation – not ideal in any environment, let alone an AI-driven one where speed and adaptability are typically critical.
Then there’s the cultural impact, where individuals operating in isolation are so focused on protecting their own corner that collaboration suffers. This, of course, makes organisations more vulnerable – with more knowledge concentrated in a few individuals, the greater the risk if those individuals leave or disengage.
Reframing AI from snake to ladder
If gatekeeping is the problem, then culture is the solution.
To overcome this, leadership needs to move from a culture of ‘knowledge is power’ to one that encourages sharing and sees that as progress. In practical terms, that means offering career paths that reward people who automate their own lower-level tasks; teaching staff to think like architects, building and auditing systems rather than just performing manual labour; and showcasing employees who have embraced AI to land higher-value opportunities.
This shift doesn’t happen overnight. It calls for a deliberate effort to recognise those who improve systems, streamline workflows, and enable others to be more effective, instead of rewarding output alone. It also calls for businesses to address the underlying fear. As research has shown us, much of the resistance to AI is around uncertainty. This has to be addressed with better communication, training, and actual examples of colleagues using AI to enhance their job, not replace it.
To use a boardgame analogy, the trick is to present AI as a ladder rather than a snake. Show people within your organisation that the most irreplaceable members of the team aren’t the ones who hold the keys to the way things used to work; it’s the people who embrace AI and other new technologies to build new ones.
This framing matters. If AI is seen as a snake – something that threatens progress and sends people backwards – then resistance is inevitable. But if it’s seen as a ladder – a tool for advancement, growth, and opportunity – then adoption becomes a natural next step.
The bottom line here is that AI isn’t going away, and neither is the pace of change it brings. But the choice really has little to do with technology itself – it is about encouraging the right mindset.
Luke is Head of Innovation at Algomarketing, where he helps teams get real value from the AI tools and technologies they use every day. He supports the AI systems Algomarketing puts in place by enabling teams through practical training, real-world use cases, and deep work inside existing workflows – redesigning them from the ground up with AI at the core.
