On constraints, coherence, and why infinite potential isn’t the same as agency.
The Persistent Idea
There’s an idea that keeps resurfacing for me — partly through reading, partly through experience, and partly through watching how large language models behave when you remove their constraints.
It’s this:
A self — by which I mean the ego or personality — is not an essence.
It’s a constrained system.
And systems without constraints don’t become free.
They become incoherent.
Infinite Potential Is Not Agency
We talk about children as “pure potential,” and there’s truth in that. Before identity hardens, almost anything seems possible.
But potential on its own doesn’t do much.
Until something narrows – a role, a value, a responsibility — nothing actually manifests.
A beam of light spread in all directions illuminates very little.
Focus it, and suddenly it cuts.
That focusing isn’t oppression.
It’s what allows something specific to exist at all.
What I Mean by “Ego”
I’m not using “ego” to mean narcissism, bravado, or a false self that needs to be destroyed.
I mean something much more mundane and much more important:
- a personality-level control system
- something that maintains continuity over time
- that filters perception
- assigns salience
- and coordinates action in a shared world
In other words, the thing that lets “you” show up tomorrow recognisably related to who you were yesterday.
That system isn’t the enemy of consciousness.
It’s the price of being someone rather than everything.
What Happens When It Dissolves
There’s a temptation, especially in spiritual or intellectual circles, to romanticise ego dissolution.
But when narrative continuity collapses entirely, people don’t become liberated.
They struggle to function.
Loss of identity coherence tends to look like:
- Fragmented story
- Runaway salience
- Pattern over-attribution
- Difficulty coordinating with shared reality
This isn’t a moral judgement, it’s a systems failure mode.
You can loosen the ego temporarily and that can be valuable but living without any stabilising structure doesn’t lead to wisdom. It leads to drift.
You can visit the void.
You can’t live there.
Why AI Makes This Visible
What’s been interesting to me is noticing how closely this maps onto AI systems.
When you remove constraints from a model:
- no role
- no task framing
- no guardrails
- no memory
…you don’t get truth.
You get hallucination.
As structure is added:
- outputs become more accurate
- tone stabilises
- personality becomes legible
- coherence improves
AI didn’t invent this dynamic, it accidentally made it observable.
The model doesn’t “have” a self but it demonstrates something important:
Coherent behaviour only appears when possibility is narrowed enough to act.
Where the Analogy Must Break
Humans are embodied. We have nervous systems, attachment, fear, pain, mortality.
The ego didn’t evolve for aesthetics, it evolved to keep an organism alive.
AI has none of that.
When a model becomes incoherent, nothing suffers. When a human does, everything is at stake – so the analogy is structural, not metaphysical.
Functional, not experiential. That distinction matters.
A Cleaner Frame
The ego isn’t an illusion to be eliminated. It’s an instrumental fiction that allows reality to be navigated.
Money is an illusion. Borders are an illusion. Laws are an illusion. They’re still real in their effects.
The self is similar. Not ultimate – but necessary.
Closing
The self isn’t what you are underneath everything. It’s what shows up when the infinite agrees to play by rules.
And that turns out not to be a betrayal of freedom, but the only way freedom ever becomes usable.
Leave a Reply