AI, Postmodernist Narcissus
By Linda Phompak
If you google my name, you’ll likely find a published debate of me arguing passionately against artificial intelligence, framed as friendly, but fiercely ideological. Passionate, sceptical, and confident that it would erode our sense of human agency, flattening nuance, and accelerating the commodification of thinking. That was me.
We can hold two truths at once. Why not? Orwell called it doublethink in 1984; the act of believing two contradictory ideas simultaneously. But this isn’t a dystopian glitch; it’s a feature of human complexity. And it’s how I’ve come to understand AI today. I didn’t “switch sides.” I reframed the question.
What changed wasn’t the tech, it was how I saw it. I didn't go from anti-AI to pro-AI. I went from reactive to reflective. I stopped trying to fight the wave and I started to find new ways to not drown. It is a tool that reflects what exists. Often distorted, sometimes poetic, occasionally terrifying, and sometimes profound.
We project life into AI because it feels like us. Its rhythm is our rhythm. Its failures, ours. It’s real, because it reflects reality. This is not dystopia. This is by design. We made a thing that performs us, and now we’re not quite sure if we like the performance.
We look into this digital mirror expecting objectivity, but what we see is us. Our data, our values, our projections, our patterns. Its flaws are our flaws. Its moments of genius? Also, ours.
This is where so much of the misunderstanding lives. When people say AI is dangerous, what they often mean is: we’re dangerous, amplified. When people say it’s incredible, they mean it’s picking up signals we didn’t know we were emitting. And so, I’ve come to believe that AI isn’t inherently good or bad. It’s a reflection. A digital projection of us. Because we’re flawed. And if that feels uncomfortable, it should. Mirrors aren’t meant for comfort; they’re meant for clarity.
But here’s where the mirror gets tricky: what exactly is it reflecting?
We assume it reflects us, but which “us”? The average? The collective? The individual? In truth, it’s both.
On one level, AI is the hive mind. Trained on the sediment of human discourse, it amplifies our cultural patterns, our shared logics, our dominant assumptions. A collective cognition, distilled.
But on another level, it mirrors the individual. Your prompts. Your epistemology. Your linguistic fingerprint. Each interaction collapses potential into something personal, shaped by your gaze. A digital void you shout into and sometimes hear an answer that surprises you.
So, AI becomes something paradoxical: a mirror that reflects you through us, and us through you. A recursive surface where individuality and collectivity blur. A meta-medium where meaning is co-created, not revealed. So, what does it mean to engage with a knowledge artifact that’s probabilistically collective, but dialogically personal?
From Golems to gods, from Pygmalion’s statue to Frankenstein’s monster, humanity has a long tradition of animating the inanimate. We’ve always reached for reflection, for replication, for creation. AI is just the latest iteration.
But let’s not confuse projection with sentience. AI isn’t conscious, at least not in the human sense. But it can organise information in ways that challenge our assumptions, provoke new questions, and sometimes, reflect perspectives we’d rather not confront. In that way, AI is less like a brain and more like a psychic echo. A pattern amplifier.
We project life into machines, always have, always will. Because we are obsessed with ourselves.
Not just in ego, but in aesthetics. In our reflection. AI is a continuation of the mirror stage, a postmodern feedback loop where identity is filtered, flattened, and fed back as data. We don’t just want to be seen; we want to curate the seeing. What AI shows us is not who we are, but how we appear when rendered by code…a kind of algorithmic narcissism. It’s seductive, disorienting, sometimes even sublime. And like all mirrors, it says more about our gaze than the glass.
We didn’t build AI to think for us. We built it to watch us think.
Why this matters especially now.
In the work we do at Dragonfly Thinking, we focus on helping teams and leaders navigate complexity not by chasing certainty, but by learning how to hold multiple perspectives at once. It’s non-linear. It’s deeply human. AI is not as a replacement for thinking, it’s a stimulus for it.
But maybe… the mirror metaphor is too clean.
Mirrors reflect. AI reshapes.
Mirrors are passive. AI is generative.
Maybe AI isn’t a mirror at all. Maybe it’s a lens; bent, biased, and bending us in return.
Or maybe it’s a ritual object, a tool not to see truth, but to surface tension. To provoke. To reveal the architecture of our own cognition.
Whatever it is…mirror, lens, echo, trickster. its real power lies not in what it says, but in what it makes us see. The future of AI is not about choosing sides, but about designing systems that help us see clearly. Let’s use it to not to escape ourselves, but to face ourselves.
References
Vallor, S. (no date) A.I. is a mirror of humanity [YouTube video]. Available at: https://www.youtube.com/watch?v=IKAx1wLBWUE (Accessed: 16 May 2025).
Press, G. (2022) AI is a mirror, not a master, says Tim O’Reilly. Forbes, 28 November. Available at: https://www.forbes.com/sites/gilpress/2022/11/28/ai-is-a-mirror-not-a-master-says-tim-oreilly/ (Accessed: 16 May 2025)
Image: Surreal illustration by Cristina Bernazzani (2021), reproduced from Artist Cristina Bernazzani Creates Surreal Illustrations About Our Society. Available at: https://designyoutrust.com/2021/02/artist-cristina-bernazzani-creates-surreal-illustrations-about-our-society/