What Do We Allow Ourselves to Know and Notice?

What AI, embarrassment and subtle perception reveal about the information we censor before we even examine it

Niina Gullstén | Executive Energy Journal | Issue #76

Welcome to The Executive Energy Journal, where we explore a side of leadership few talk about —the subtle energy we have, live in, and how it impacts clarity, decision-making, and performance.

How EEJ works

What's measurable: what research and observable mechanisms support
What's unknown: where evidence ends, and interpretation begins
What I do pragmatically anyway: the simplest reliable practice I use with clients
Your role: take what is useful and leave the rest

In this issue:

• Why we may censor unusual perceptions before we ever put them into words
• How AI can both ground and narrow the way we interpret intuition and energy
• Why experience, interpretation and mechanism need to remain separate questions

I came across an interesting MIT Sloan study this week.

The researchers were looking at whether people would rather ask a human or AI for professional advice. In general, people preferred the human and considered humans more competent.

Then the researchers made the problem embarrassing, and the preference shifted towards AI.

In one of the experiments, participants described a real, embarrassing financial or technology-related problem to either a human or an AI adviser. They justified their behaviour much more often when they thought they were speaking to a human: 33% of the time, compared with 15% when speaking to AI.

I can understand that very well.

One of the useful qualities of AI is that you can put something into words without simultaneously managing another person's response to it. You can admit something, turn it around, examine it from different directions and see what happens. I've done a lot of that myself.

And because much of my work involves intuition, subtle information and energy, I have also noticed something else.

1. What AI does with unusual information

AI can be extremely useful for this work.

I can describe something I perceive and ask AI to separate my observation from my interpretation. I can ask for alternative explanations, or what neuroscience, psychology or another research field might already explain. I can test an idea against what is known and identify where the known science runs out.

That grounding function and its speed are very valuable to me.

At the same time, I have learned that if I simply let the conversation follow its default direction, subtle energetic observations in particular can gradually get translated into something more conventional.

An energetic perception /intuition may be reframed as a nervous-system response, projection, pattern recognition, hypervigilance, symbolism, an emotional reaction, or some combination of those. The conversation may me skewed towards "which explanation is the correct one." For me, that tends to not be the most useful question. 

Because we perceive and interpret with our full beings.

The nervous system may be responding. The body may be registering something. Memory and psychology may be shaping the interpretation. Pattern recognition may be involved. And there may also be an intuitive or energetic perception in the mix.

Those layers are not mutually exclusive.

In many cases, I suspect the mechanism is precisely an interaction between them.

What matters is not reducing the experience to the most conventional explanation available, but examining the different layers carefully enough to understand what is actually happening.

If I say that I feel something in the air in a room, the first piece of information is that I feel something in the air in a room.

I may eventually conclude that the experience came from my own nervous system. I may find behavioural information that supports what I sensed. I may decide I misread it entirely. I may continue to experience it as energetic information for which I do not have a good explanation that would be grounded in science.

I want all of those possibilities available while I investigate.

And this is where that MIT study sent my thinking in another direction:  Before I can tell AI what I have perceived, I have to allow myself to perceive it.

2. The information that never reaches the conversation

We tend to think censorship happens when we decide whether to tell somebody something, but a surprising amount of it happens earlier.

We have internal rules about the kinds of information we consider respectable enough to notice.

I grew up in Finland, where my experience of the dominant culture has been strongly analytical and engineering-oriented. There are many advantages to that, and I respect and even enjoy a lot of it. Precision and evidence matter. Making claims you cannot substantiate is generally discouraged.

There can also be a fairly narrow doorway through which an experience has to pass before we are comfortable calling it information.

That produces an interesting contradiction, because Finns do privately hold beliefs and experiences that sit outside a straightforward materialist model.

A 2024 study looked specifically at 387 Finns who were oriented towards science. The researchers still found both religious and more secular supernatural explanations among them, and some participants actively integrated supernatural causality with scientific explanations.

People are apparently quite capable of carrying more than one explanatory model. Whether they feel equally comfortable speaking from both of them is another question.

Living for almost eleven years in Southern European and Latin environments gave me a different experience. Intuitive information seemed more socially ordinary. People could know something, sense somebody's intentions, refer to an atmosphere or simply say that something felt wrong without needing to construct a defence of the statement first.

What struck me there was something different: the experience was often accepted without being examined very far.

In North American contexts, particularly coaching and personal-development ones, I have often encountered extraordinary fluency around energy. People will readily describe what they feel in a room, what they are picking up from someone, where energy is moving or what feels blocked.

The bridge into serious, rigorous scientific inquiry seems less frequently built, at least publicly.

Those are observations from my own life rather than a cultural taxonomy. But living between these environments has made me notice how much culture influences the point at which we interrupt our own perception.

3. Experience can arrive before explanation

This is interesting to me because I suspect that a great deal of phenomena still described as supernatural will eventually turn out to have perfectly natural mechanisms that we simply do not understand yet.

"Natural" is an expanding category.

Magnetoreception is one example I find fascinating.

A wide range of animals can detect the Earth's magnetic field. As far as I know, the sense is well established across major vertebrate groups and some invertebrates, although the underlying mechanisms are still being investigated.

In a 2019 human study researchers exposed participants to controlled rotations of Earth-strength magnetic fields and found specific, repeatable changes in alpha-band brain activity. The authors concluded that at least some human brains were transducing changes in geomagnetic fields into a neural response. Whether this produces a usable conscious human magnetic sense remains unresolved, and the study debated.

I am not suggesting that magnetoreception explains subtle energy perception. I use it because it is a useful reminder of how strange the sensory world can look before we understand the mechanism.

The boundary around what a human being can perceive is therefore a reasonable subject for inquiry.

4. Three separate questions

I find it useful to keep three questions apart.

1. What did I experience?
2. What do I think it means?
3. How is the experience produced?

We often rush from the first question to the third.

• If we cannot explain the mechanism, we start distrusting the experience.
• Or we go in the other direction and become so certain about the meaning of an unusual experience that we stop investigating it.

Neither helps much with discernment.

If I perceive a dense, unpleasant quality around a room, I can allow myself to register that accurately.

I can then ask what else I noticed. What happened in my body. Whether the feeling changes when I leave the room. Whether it is familiar. Whether I have been accurate when I have sensed this quality before. Whether their behaviour supports any of the possible interpretations. Whether there is something in me that could be generating the response.

And I can leave the mechanism open.

That gives me far more information to work with than immediately correcting myself into, "I suppose I just don't like this group of people."

5. So, what do we allow ourselves to know?

Let's go back to looking at AI.

The MIT study suggests that reduced fear of judgment can make us more willing to disclose information. That could be particularly useful for perceptions we have learned to consider embarrassing, irrational or professionally unacceptable.

You can tell an AI:

Something changed in the room when a particular person started speaking. I felt it before anything obvious happened.

Or:

I know this sounds strange, but the space around this person feels different to me.

Or:

I have known for three months that this project is wrong, and I cannot tell you how I know.

Then you can examine it.

You can ask for conventional explanations. You can test your interpretation. You can look for evidence. You can ask what would disconfirm your conclusion. You can compare the experience with what happened later.

You can also tell the AI very explicitly not to erase the observation merely because there is currently no accepted scientific explanation for the mechanism you think you may be perceiving.

That last part is notable here.

AI has access to an extraordinary amount of human knowledge, so it will reflect many of the implicit and explicit assumptions contained in that knowledge. Used passively, it can reinforce the boundary of what is already considered reasonable.

Used deliberately, it can help us examine the territory around that boundary.

But there is one thing it cannot do for us.

It cannot examine information we have already refused to let ourselves know.

So perhaps before asking whether your intuition is right, whether subtle energy is scientifically real, or what AI makes of what you are experiencing, there is an earlier question:

What do I already perceive that I keep correcting myself for perceiving?

There may be nothing dramatic there, just information.

And once it is allowed into awareness, you can start finding out what it actually is.

Until next week,

Niina

P.S. Last Friday's Debrief is a short experiment with exactly this: allowing yourself to sense what you already know, within you and in the space around you, before trying to interpret it. If you subscribe by 20 September, you'll get that issue as well. 

Friday Debrief subscription

References

• So, E., Sussman, A., & Yang, F. Y. AI Advisers and the Competence-Judgment Tradeoff in Information Disclosure. Research discussed by MIT Sloan, 31 August 2026.
• Haimila, R., Metsähinen, H., & Sevalnev, M. (2024). Supernatural Belief in ‘Scientific’ Worldviews? Journal of Cognition and Culture, 24(1–2), 1–34.
• Wang, C. X., Hilburn, I. A., Wu, D.-A., et al. (2019). Transduction of the Geomagnetic Field as Evidenced from Alpha-Band Activity in the Human Brain. eNeuro, 6(2).