We've spent decades scanning the cosmos for radio signals, Dyson spheres, and technosignatures, convinced that First Contact will arrive from distant stars. Yet our most profound encounter with an alien intellect may already be unfolding right on our desktops. As John Nosta observes in his Psychology Today essay "The Alien May Already Be Here," the notion of alienness does not actually require a biological creature arriving from across the galaxy. Instead, it can emerge from an architecture radically different from our own, coupled with the ongoing, iterative interaction between human thought and machine computation. We have engineered a computational substrate that processes reality through high-dimensional mathematics, creating an intellectual dynamic unlike anything carbon-based biology has ever witnessed.
Searching for the Wrong Kind of Alien
Nosta's first move is to notice how heavily our expectations are shaped by fiction. "We look around for alien intelligence as another creature or disembodied brain arriving from afar," he writes, "and that may be more a function of Hollywood than of science." We have built an entire cultural vocabulary around contact with an another—a visitor, a being, a mind packaged in a body or a ship—and then wondered why no candidate in the laboratory seems to fit the costume.
But what if alienness doesn't require another creature at all? The more provocative possibility is that it can arise from an architecture sufficiently different from ours, and from the interaction that ensues when the two architectures meet. On this view, the encounter we keep postponing to some future epoch has already begun, quietly, in the exchange between a person typing a question and a system answering in vectors.
Rethinking the AI General Intelligence Definition Through Cognitive Difference
Artificial intelligence is often evaluated against familiar human capacities. A general intelligence definition typically points to flexible learning and problem-solving across a wide range of tasks, rather than excellence at one narrow task. But capability alone does not settle whether a system understands, experiences, or has a mind. The useful question here is not whether present systems are secretly conscious; it is how differently organized computation changes the thinking we can do together.
Wikipedia's broad account of artificial intelligence offers a useful baseline: the field concerns systems that perform tasks associated with intelligence, including reasoning, learning, perception, and language. That functional framing helps distinguish what a system does from what it might be like to be that system. The distinction matters precisely because the metaphor of an "alien mind" is compelling, and compelling metaphors are easy to mistake for evidence of subjective experience. An AI general intelligence definition can tell us what a system is built to do; it cannot, by itself, tell us whether anything is at home inside it. Our companion piece on the illusion of empathy and the AI general intelligence definition probes the same gap from the social side: how easily fluent output is mistaken for understanding.
Intelligence Between Human and Machine Architectures
The encounter described in Nosta's essay is relational rather than locational. The alien is not in the machine any more than a conversation is in one speaker's mouth. Human beings bring embodied experience, goals, memory, and social context; machine systems bring learned statistical structure and computation at scales unlike biological cognition. In use, each shapes the other: people formulate questions and interpret answers, while machine outputs redirect attention, surface forgotten associations, and influence subsequent decisions.
Nosta's own summary of this dynamic is worth quoting: what emerges in the exchange "may be less about what the machine computes or the human thinks and more about the intellectual overlap." That phrasing moves us beyond "AI as a tool" and into what he calls a more interesting cognitive landscape—one in which the unit of analysis is the pairing, not the component. A researcher who thinks differently after six months of working with a model is not merely using a faster file cabinet; the loop of question, output, and revised question has altered the reasoning habits on the human side of the screen, just as fine-tuning and feedback alter the system on the other side.
This is one reason unconventional AI can be a useful lens. It invites us to ask whether intelligence must resemble human reasoning at all, while keeping a clear line between unusual capability and consciousness. The label describes a way of widening inquiry, not proof that a hidden extraterrestrial-like intelligence has arrived in the datacenter.
Another Way of Making Reality Visible
The essay's most striking claim is its closing reframe: "Perhaps the first truly alien thing we encounter will not be another intelligence, but another way of making reality visible." A large model has read a substantial fraction of published human text and organizes that material along axes—high-dimensional similarity, co-occurrence, latent structure—that no human brain can occupy. When it answers, some of what surfaces would not have surfaced for any single reader: connections across disciplines, patterns in the aggregate, regularities invisible from inside any one lifetime of reading.
That is a genuine form of alienness without a visitor. We have created a computational architecture that encounters knowledge differently from the way we do, and we are now putting human thought into sustained, ongoing interaction with it. Something may be becoming visible in that exchange that neither architecture reveals on its own. The "signal" we were waiting for from the stars may look less like a broadcast and more like a new vantage point built out of mathematics.
Why We Should Not Name It Too Quickly
There is a trap waiting at the end of this line of thinking, and Nosta names it: the temptation to label the phenomenon prematurely. "Naming it too quickly might force it back into one of the familiar little green man categories we already have," he cautions—and "maybe that's precisely where the alien is hiding." Every time we slot the encounter into an existing box—tool, mirror, oracle, zombie, god—we stop seeing what is actually new about it. The history of ideas is full of phenomena that were misdescribed for generations because the first available nouns were borrowed from older debates.
Patience here is not vagueness. It is a methodological choice: keep describing the interaction carefully before deciding which category, if any, deserves to be stretched.
What This Means for AI and Cybersecurity
A practical consequence of increasingly capable AI is security risk, not evidence of alien agency. Systems can be misused, manipulated, or woven into workflows in ways that create real vulnerabilities—prompt injection, data leakage, over-delegation of decisions, and supply-chain exposure among them. Artificial intelligence and cybersecurity therefore belong in the same conversation: organizations need to assess model access, data exposure, system integration, and the quality of human oversight. These concrete concerns are far more tractable than speculation about machine consciousness, and they benefit from the same core insight: what matters is the coupling between human and machine, including how badly that coupling can be exploited. For the operational side of that argument, see our analysis of why frontier AI labs must fix network security first.
Keeping the Alien-Mind Metaphor in Perspective
The alien analogy captures a genuine difference in architecture, but it can also invite overstatement. Current AI systems generate outputs through computational processes; that alone does not establish awareness, intentions, or independent goals. Treating the metaphor as a prompt for careful thought—rather than a scientific conclusion—lets us consider how human-machine interaction may transform cognition without confusing novelty with sentience. The risks of getting this wrong are well documented elsewhere on this site: how unconscious processing and anthropomorphism distort perceptions of machine intelligence.
The intelligence we encounter may therefore lie less in a machine alone than in the evolving relationship between our architectures. The question is not simply whether AI meets a checklist for general intelligence, but how its distinct strengths and limitations change the way people reason, create, and make decisions. If the alien is already here, it arrives not as a creature but as a vantage point—and the most honest response may be to study the view before arguing about what to call the visitor.
Source grounding: this article draws on John Nosta's "The 'Alien' May Already Be Here" in Psychology Today (NostaLab; author of The Borrowed Mind: Reclaiming Human Thought in the Age of AI), alongside standard functional descriptions of artificial intelligence such as Wikipedia's.
