The Mind That Holds More
In one of my recent coaching conversations, a more junior leader asked me: "I am observing my senior leaders and what I am always impressed about is how they see something I don't see. I see only one side of the picture, and they can see both, and even more. How does one develop this skill?"
It is a great question. And the honest answer is: slowly, over time and only through doing the work yourself.
To illustrate this point, I can think of another leader I worked with for some time. When questions arose about his team's engagement – team’s frustration about decreased funding in new products and his own view of their declining output as a team - he could see his team's perspective: team felt deprioritized due to reduction of resources.
Simultaneously, he could see the company's perspective: a strategic shift toward shifting business priorities and product offering was driving resourcing decisions across the entire business.
He also could clearly see the reasons behind the mandate to focus on acquiring large customers over anything else, which his team on the other hand, translated as a short-term thinking. The senior leader held both perspectives. He did not turn the complexity and the tension coming from his team into a simple story. He navigated the polarity and still moved forward. This is an example of an expanded capacity to hold complexity.
This skill is built, not given.
Cognitive psychologist Gary Klein spent twenty years studying how experts - firefighters, military commanders, surgeons - make decisions under pressure. He found that experts don't carefully deliberate between options. They read the situation and they know what to do, drawn from their real world experience after having made thousands of real decisions in high-stakes moments.
He called this Recognition-Primed Decision Making and it is one of the most influential models in the field of decision making to date. For an outsider it looks like these experts are using intuition, the gut feeling. However as Gary Klein described the process, it is based on four specific sources of expert power: intuition, mental simulation, the use of metaphor and analogy, and storytelling.
On the contrary, less experienced individuals have to think the problem through from scratch every time - comparing options, working harder, often for worse results. It is over time that their decision making ability improves.
Interestingly enough, it is the same technique that Navy SEALs use before high-stakes operations. They call it mental rehearsal. Before entering a mission, they walk through every scenario in their minds: what they'll encounter, what and where decisions will need to be made, what could go wrong and how they'll respond. It's deliberate cognitive preparation.
Mental rehearsal builds that experience to draw from before the crisis arrives. Navy Seals are giving their brain "experience" it hasn't lived yet. When the real moment comes, their nervous system says: "I've been here." Being then able to recognise the situation, they are able to move fast and make high-quality decisions, even under extreme cognitive load.
In his book “Sources of Power” G. Klein shares a story. A fire lieutenant sent his crew into a burning house. They sprayed water, the fire roared back. Something felt wrong, though he had no specific evidence. He ordered his men out. The floor immediately collapsed. Turned out there was an empty basement below which they were not aware of. However, it was his years of experience that gave him an internal signal to make this decision because something felt off. A less experienced commander would have asked their men to stay inside.
He also found that in 16 out of 25 decisions with poor outcomes (in some occasion fatal ones), the cause was simply lack of experience. Not poor reasoning. The person had never seen this before and didn't know what they didn't know.
This is exactly what your junior leader is watching when he sees senior leaders perceive what he cannot. It is not intelligence. It is an accumulated source of knowledge and experience - built decision by decision, consequence by consequence, over years.
Which brings me to AI.
With AI, advice is available at the reach of a finger, in seconds. And instead of using our own judgement and validating with other resources, we find ourselves outsourcing our thinking entirely. Instead of AI being a tool, it becomes the brain and we follow its direction without scrutiny. Prior to AI this would look like speaking to someone more experienced, consulting management books, learning from mentors, or engaging in further study. Those routes were slower. They were also formative.
Researchers Steven D. Shaw and Gideon Nave from Wharton School of the University of Pennsylvania calls this “cognitive surrender” in their recent landmark paper. What they found that the 3rd way to make decisions in todays work environment is not anymore using AI as a tool – it is to outsource the entire decision making and critical evaluation to AI.
Across 10,000 trials, participants who used AI performed 25 points better when AI was right and 15 points ‘worse than their own unaided thinking’ when AI was wrong. They couldn't tell the difference. Their confidence went up either way. From research point of view this difference is significant.
What is critical in this study is that it was not that people were not just getting answers wrong, they were getting them wrong with higher confidence than they would have had thinking alone. They felt more certain precisely because they had stopped thinking for themselves and outsourced their thinking to AI.
The research also found that the people most vulnerable to cognitive surrender were those with higher trust in AI and lower drive for effortful thinking. Those with higher fluid intelligence were most resistant, had stronger analytical capacity and genuine intellectual curiosity.
In other words, those who had developed the habit of sitting with difficulty and challenge, of not reaching for the easy answer, were the ones who could still think clearly when AI got it wrong.
The Workslop
I recently spoke with a senior leader who had started to notice something troubling. His more junior team members were submitting long, polished documents for review. But as he read carefully and dug in, he realised none of it reflected their actual thinking. It was AI-generated.
A recent HBR article named this phenomenon the ‘workslop’ - work that looks sophisticated but contains no real cognitive effort from the person who submitted it.
With my client, we spent time together thinking through how to address it. How to hold honest, challenging conversations with the team. How to invite people to share shorter, rougher, genuinely personal work , even unpolished , because what matters is that the thinking is theirs. The instinct is understandable: AI smooths your grammar, sharpens your structure, offers solutions with confidence. And as Shaw and Nave's research shows, more junior people tend to trust it more than they trust themselves.
The cost for organisations.
Klein's research tells us that expertise develops through a very specific process: real decisions, real consequences, real feedback, repeated across increasingly complex situations. The junior leader submitting an AI-generated document is not making a judgement call and living with the outcome. They are not building the pattern. The decision muscle goes unexercised. And unlike a physical muscle, you cannot always feel the it weakening.
What concerns me most is not the present performance, it is the future capacity. The senior leader your junior coachee admires, the one who can hold both perspectives and still move forward, developed that capacity over years of sitting with difficult decisions without a shortcut available. If that developmental window is bypassed, what will leadership look like in ten years?
This expanded cognitive capacity is what many senior leaders today have that more junior people are still developing. This is not about high IQ. High technical expertise. It is about the ability to hold more, see more, and still decide.
And that capacity (ideally) should not be outsourced.
The Invitation
The question I am sitting with is this: where in your work are you currently letting AI think for you, when the harder and more important thing is to form your own view?
AI is a remarkable tool. Used well, it can extend what we are capable of. But the pattern recognition library , the invisible accumulation that makes a senior leader see what others cannot, builds only one way. Through sitting with complexity. Holding the tension. Making the judgement yourself. Watching what happens. And taking that insight and event forward into the next decision.
References:
Shaw, S. D., & Nave, G. (2026). Thinking—fast, slow, and artificial: How AI is reshaping human reasoning and the rise of cognitive surrender (Version v1) [Preprint]. OSF. https://doi.org/10.31234/osf.io/yk25n_v1
Klein, G. A. (1998). Sources of power: How people make decisions. MIT Press.
Niederhoffer, K., Rosen Kellerman, G., Lee, A., Liebscher, A., Rapuano, K., & Hancock, J. T. (n.d.). AI-generated “workslop” is destroying productivity. Harvard Business Review. https://hbr.org