
There is a clear, crisp vision of how AI could work in professional settings. It can remove the boring, monotonous tasks that take up so much of our time at work, freeing up time for us to be creative and focus on the harder end of problems that require original thought. It is already becoming established doctrine that there are crucial tasks and responsibilities where AI will require a human in the loop – tasks that are particularly complex, high-stakes, or need human connection. This vision creates a more meaningful experience at work and in life. We are spending more time on the core business of being human and potentially have more time with our loved ones or in our chosen pursuits.Â
Here’s the paradox: unless we are deliberate and careful in the way we use AI, it can erode our ability to be that top-tier human thinker, doing the things that only humans can. Just as AI comes along promising more time for the strategic and creative problems you struggle to get to, it can also rob you of your ability to tackle them. This isn’t great for AI, either. Good outcomes for AI will often depend on the human–AI partnership, but without being very deliberate, partnering with AI may erode your ability to be the human end of that partnership.Â
Integrating AI is a tightrope walk, with a significant fall at stake. Make it across the chasm and we’ll find the potential to lean into our unique, embodied, creative selves. Misstep and we might lose the gamble – and our ability to think as well as we could.Â
The interplay between the brain’s three core networksÂ
What’s going on in our brains that allows us to dismiss tasks once they are complete? It’s an interplay between three core networks that coordinate how we think and feel. Each one has a different job:Â
- The Central Executive Network (CEN) – as your task-focused brain – handles working memory, attention and executive functions when you’re actively solving problems. I imagine I can feel this one firing when I hurtle through a busy workday.Â
- The Default Mode Network (DMN) – as your reflective brain – activates when you’re daydreaming, remembering the past, imagining the future and thinking about yourself and others. That’s where you ponder, think more deeply and generate creative, complex solutions.Â
- The Salience Network (SN) – as a switch in your brain – determines when you focus outwards on tasks and when you turn inwards for reflection. I imagine it as an old-fashioned railway signalman, switching the tracks for my mind to run along depending on the desired destination. These three networks work together – although not always at the same time – to create the high-level thinking and emotional behaviours that make us human.Â
Critically, the CEN and DMN are typically anticorrelated: when one turns on, the other turns off. Â
How AI causes cognitive debtÂ
Keep these networks in mind as we jump to 2025. A study from MIT goes viral with an alarming headline in The Telegraph: ‘Using AI makes you stupid.’ The research reveals something crucial about what happens when we use AI through prompts. When we ask AI to complete a task for us, the CEN of the brain treats it as complete immediately. The brain’s executive network concludes its work the moment the AI generates a response. Here’s the critical part: the railway signalman of the SN never switches up the tracks to involve the deeper thinking network. The task never passes over to the DMN for deeper processing – for pondering, wondering, integrating with existing knowledge or making meaning.Â
MIT researchers found that students using ChatGPT displayed significantly reduced brain activity in areas linked to attention, memory and advanced reasoning compared to those who wrote independently or used search engines. More concerning, those who had used AI performed worse on subsequent tasks and struggled when asked to work without it. The researchers introduced the term ‘cognitive debt’ to describe the cumulative toll: similar to how using satnav has eroded our sense of direction. Cognitive debt is akin to ‘technical debt’ in software engineering: there is a short-term gain but a hidden liability is created. In this case, you get an easier and quicker answer, but long term you are losing your mental fitness to think critically, meaning that you will need to put in more effort to recover that lost cognitive ground.
A separate study from the University of Pennsylvania, involving over 4,500 participants, found that those who used ChatGPT to research topics ‘developed shallower knowledge’ than those using Google Search. Their knowledge was less invested, sparser and not as original – precisely because the information synthesis was done for them rather than actively constructed through their own cognitive effort.Â
The opportunity we might squanderÂ
AI genuinely offers the promise to automate tasks that should be automated: the repetitive, the mundane, the purely mechanical. This could free human capacity for what industrialisation suppressed: work that allows meaning-making, growth and development towards mature adulthood. Emotional maturity in the workplace – characterised by empathy, acceptance of responsibility, openness to feedback and learning, and adaptation to change – enables not just professional success but a natural maturation for the human mind. Through work experiences, we can progress through increasingly sophisticated levels of thinking and meaning-making.Â
Meaning-making is taking in everything in your experience and creating a way of working with it. It is how you make sense of what is happening – the story you tell yourself about it – and that underpins how you choose to move forward. Â
For example, if a colleague at work gives you some blunt feedback, the words are part of your experience but so too are the tone of voice, the environment you are in, your physiological response and how it connects to the broader context of your work and organisation. Your experience shapes your beliefs, and your beliefs determine your actions. How you process your experience is the meaning-making.Â
Preserving our capacity for thinking, reasoning, and meaning-making requires deliberate intent: ensuring technology remains a partner to human effort rather than a replacement for it.Â



