Artificial intelligence is fundamentally transforming how we complete tasks. In AI-first companies, a new work reality is emerging: people at all levels are taking on coordination and control tasks when it comes to collaborating with AI systems. They’re effectively becoming managers – of parallel AI processes, of selecting the right tool for each task, of quality assurance. And this, even though they carry no formal leadership responsibility.
The question isn’t whether this change is coming. In many cases, it has already happened. The question is: Are we aware of this – and are we equipping our teams and individuals with the right competencies?
The Creeping Change in How We Work
In the past, the world of work and the required skills were manageable: One task, one tool, one result.
Today, reality looks different. A typical morning in an AI-integrated company might look like this:
- NotebookLM creates research syntheses from various sources
- Claude Cowork organizes files and assigns tasks
- ChatGPT summarizes meeting notes and identifies associated action items
- Midjourney generates visualizations for a presentation
Another difference from the past: All of this doesn’t happen sequentially, but in parallel. Different AI systems work simultaneously on different subtasks – and someone needs to keep track, prioritize, ensure quality, and above all decide which AI tool is best suited for which task.
This coordinating role has nothing to do with position in the organizational chart. It emerges from the changing way of working itself, which unfolds on two levels: first with individuals, then in teams. Both levels require new competencies – and the necessary depth of these additional skills is often underestimated.
AI-First transformation means shaping change deliberately. It’s not about having perfect solutions. It’s about shaping change with intention, rather than letting it happen by default.
Level One: Individual Contributors as AI Coordinators
The first change affects everyone individually. Anyone working productively with AI today must be capable of managing multiple tools, ideally running in parallel. This requires additional skills that we previously expected primarily from managers or project leaders:
Prioritization between parallel processes: Which AI task is more important right now? What’s running in the background, which task requires active control or iterative interactions with the AI?
Quality control of outputs: AI delivers results – but not always the right ones. Responsibility for quality remains with humans. This means: critically reviewing, adjusting, and iterating again.
Tool selection and evaluation: Which system is suitable for which task? NotebookLM for complex research, Claude for context-intensive text work, ChatGPT for quick summaries? It’s not enough to simply make “all” tools available. Every employee must also be able to decide which tool is best suited for which task. This means that informal knowledge within the team – often called “tribal knowledge” – becomes a critical success factor: active exchange among colleagues where they can constructively share their experiences from various experiments with different tools.
Context switching: Switching between different AI interfaces, different tasks, different ways of thinking. This is cognitively demanding and requires conscious structuring.
Maintaining overview: Perhaps the biggest challenge. When four different AI processes are running in parallel – where am I right now? What’s the next step? What’s still missing?
All of this is, in a sense, management work. We just don’t call it that.
AI First: New Competencies on Two Levels
AI integration is fundamentally changing the way we work. Every employee becomes a coordinator of parallel AI processes — at both the individual and team level.
Level Two: Team-Level AI Coordination
The second, more complex level emerges when not just individuals work with AI, but entire teams use various AI systems in a coordinated way. This is where it gets really challenging.
An example from our own practice at NordAGI: When we work on a more comprehensive project, the division of tasks might look like this:
- Person A conducts research – partly with NotebookLM, partly with Claude, depending on which sources are available and how complex the synthesis needs to be. Often it’s even a combination of various tools being used, whose results then need to be merged.
- Person B creates a whitepaper draft based on this research with Claude, because the extended context and iteration are particularly valuable here.
- Person C develops infographics or visualizations with various AI tools, depending on requirements and the desired end format or channel through which these visualizations will be made available (blog post, LinkedIn, other social media channels).
On one hand, this sounds like classic division of labor. The difference is: Each person doesn’t work with just one tool, but coordinates multiple AI systems themselves. At the same time, outputs need to be aligned with each other. Coordination happens on two levels: between team members and the various AI systems being used in parallel.
In our 5-stage model of AI integration, we describe this transition as a critical development step: from individual AI use to systematic team integration. Many companies underestimate what new demands this step places on coordination, communication, and skills.
This form of collaboration between humans and artificial intelligence – and simultaneously between teams and various AI systems – is a new work format for which established patterns often don’t yet exist, which is why we “navigate by sight.”
The Underestimated Complexity
One reason why the challenge is often underestimated lies in the AI tools themselves. They appear so accessible that it hardly seems necessary to actually develop skills. After all, “anyone can use ChatGPT” – that’s basically true, albeit with varying degrees of success. But coordinating multiple AI systems in parallel, knowing their strengths and weaknesses, critically evaluating outputs, and using them in a coordinated way within a team? That’s a different level of complexity.
Added to this: these skills don’t develop automatically. People learn them through experience, through trial and error, through exchange with colleagues – including through the tribal knowledge that emerges within teams.
This Learning Process Requires Time, Space, and Conscious Support
And the understanding that it extends across all hierarchical levels. A middle manager today might work in three different working groups, use AI tools themselves for their own tasks, and simultaneously need to support their team with AI integration. They are thus both a specialist without leadership responsibility and a team member in various contexts, and beyond that, also a leader.
The Requirements Multiply
The Leadership Task: Creating Awareness, Enabling Skills
When the way of working changes so fundamentally, new questions arise for leaders:
- What competencies do your employees need to work productively with various AI systems? Is it just about tool knowledge – or about fundamental capabilities like prioritization, quality control, critical evaluation of AI outputs?
- How do you enable your team to take on this coordinating role? Is there room to experiment? Can employees try out different tools and figure out what works for what purpose?
- How does your team coordinate AI usage? Does everyone work independently, or are there conscious agreements about who uses which systems for which tasks? Are experiences shared – is tribal knowledge actively promoted?
From these reflections, concrete starting points for practice emerge. Leaders should create the following framework conditions:
Time and space for competency development: Basic AI skills might be acquired on the side. But the competencies for professional, productive AI work don’t develop on the side. They need time, space, and necessary support: the freedom to experiment, gain experience, fail, readjust – and exchange within the team and beyond.
Exchange and knowledge transfer: The best learning source is often colleagues’ experiences. Create formats where teams can share their AI working methods – informally and without pressure. This is where the tribal knowledge emerges that is indispensable for successful AI integration.
Clarity about responsibility: When employees produce AI outputs, it must be clear: Who bears responsibility for quality and correctness? This cannot lie with the AI – it lies with humans. This responsibility must be consciously assumed.
Systematic approach instead of wild growth: AI integration needs structure. Not necessarily strict rules, but shared understandings: Which tools do we use for what? How do we handle sensitive data? How do we document AI-generated content? And above all: How do we map our processes with AI tools?
AI-First Transformation Means Consciously Shaping Change
It’s not about having perfect solutions. It’s about consciously shaping change, rather than just letting it happen.
AI First Is More Than Tools
The transformation to an AI-first company doesn’t mean giving every employee access to ChatGPT. It means establishing a new way of working in which AI systems are naturally integrated – and in which all employees have the competencies to work productively with these systems.
This requires new skills. Coordination, quality control, tool evaluation, prioritization, maintaining overview. Skills we previously expected primarily from people in management functions. Today, everyone who works with AI needs them.
Would you like to strategically implement AI in your daily business operations?
The question is: What does this look like in your company? Do your employees have the competencies they need for this new work reality? Or are AI integration and competency development currently happening side by side – without systematic connection?
In a no-obligation strategy session, we would be happy to introduce you to the NordAGI approach.
AI Makes Us More Productive – And That's Precisely the Danger | NordAGI
[…] our article “AI First Means: New Skills for Everyone”, we explored the additional competencies that emerge when people work productively with AI: […]
AI Workflows for Practice - How AI Becomes a Team Member
[…] It sounds paradox: AI promises to take over routine tasks, optimize workflows and generally lighten our workload. In an earlier post, we explored why the introduction of AI requires all employees to develop new competencies — including those who have never held management responsibilities (“AI First Means: New Skills for Everyone”). […]
AI Strategy for SMEs: Ideas for 2026
[…] transparent communication, genuine involvement, and sufficient time for capability development. AI demands new skills from everyone — from senior leaders to administrative […]
How is AI rewriting the headcount vs value creation equation? | NordAGI
[…] capabilities that must first be developed — a challenge explored in detail in our article AI First means new skills: why all employees need to understand AI management. According to a 2026 Deloitte study, only 33 per cent of organisations are actively redesigning […]