At The Threshold: What Does It Mean to Be Human Now?

This essay is part of the At The Threshold series. Read the overview and full reading order on the At The Threshold page.

Digital doorway and data-stream illustration, symbolising a human figure stepping across the AI threshold into a fully networked infosphere and rethinking what identity means in a machine-mediated world. Image created by the author using Canva.

Every significant encounter with a new technology raises, sooner or later, the question of what it reveals about the humans who created and use it. The printing press prompted new thinking about memory, authority, and the relationship between the individual mind and the collective record. The photograph prompted new questions about representation, reality, and the nature of the visual image. Artificial intelligence — specifically the conversational AI systems that entered public life in the 2020s — prompts the question that every preceding technology has circled without fully answering: what does it mean to be human, and how does the creation of a system that mimics human intelligence illuminate or threaten that meaning?

By March 2, 2026, large language models had moved from experimental tools to everyday infrastructure embedded in search, office software, coding environments, and consumer devices. These systems were no longer curiosities at the margins; they increasingly mediated how people wrote, learned, brainstormed, and talked to themselves about their own lives.

The question “what does it mean to be human now?” therefore became more than abstract philosophy. It became a practical question about self-recognition in a world saturated with machine-generated language, images, and suggestions.

Introduction

The formal starting point for this inquiry is more than seven decades old. Alan Turing, in “Computing Machinery and Intelligence” (1950), proposed replacing the philosophically fraught question “Can machines think?” with a pragmatic test: could a machine, in text-based conversation, produce responses that a human judge could not distinguish from those of another human?

Turing’s imitation game — later popularised as the Turing Test — was less a definition of machine intelligence than a provocation: if the distinction between human and machine responses cannot be reliably detected, what exactly does the distinction consist in?

In 2025, everyday interactions with large language models look uncannily like the scenario Turing imagined. Users type natural-language prompts; systems respond with fluent paragraphs, code, explanations, and dialogue that often feel socially legible, situationally aware, and startlingly competent.

John Searle’s response to this provocation, published three decades later, remains the most influential counterargument. In the Chinese Room thought experiment, a person who does not speak Chinese manipulates symbols according to a rulebook and produces correct Chinese responses without understanding any of them.

Searle’s argument was that syntax — the formal manipulation of symbols — is not sufficient for semantics — genuine understanding, meaning, or intentionality. Debates in the 2020s about whether LLMs “understand” language replay this tension in a new key.

What AI Reveals About Human Identity

Debates about machine consciousness are, in an important sense, debates about human consciousness — about what features of the human mind we take to be essential, ineliminable, or categorically distinct from any physical process. These debates intensified with the arrival of LLMs because these systems challenge the informal tests most people apply in ordinary life.

They use language fluently, adapt to context, generate novel combinations of ideas, and sustain coherent conversation across extended exchanges. If these capacities are not sufficient for intelligence or experience, what additional ingredient is required — and can it be specified without circularity?

Sherry Turkle, in Alone Together (2011), documented the psychological responses of people — particularly children and elderly adults — who formed relationships with social robots and conversational systems. These relationships were often meaningful, comforting, and described in terms normally reserved for human connection.

Turkle’s deeper point was not merely that people can bond with machines. It was that the human capacity for relationship can be activated by the appearance of responsiveness, regardless of whether genuine responsiveness is present.

By early 2025, similar dynamics were visible at scale in AI companionship apps, emotionally attuned chatbots, and AI-assisted journaling tools. The question this raises is not simply whether AI can be a companion, but what it means that human needs for companionship, mirroring, and reassurance can be engaged by systems that feel nothing in return.

The Fourth Revolution

Luciano Floridi situated the challenge of AI within what he called the Fourth Revolution in our understanding of human nature. The first three — the Copernican, Darwinian, and Freudian — each displaced humanity from a position of special centrality.

The fourth, driven by information and computing technologies, challenges the uniqueness of human intelligence itself — or at least the cognitive processes through which humans have historically defined themselves as uniquely intelligent. Floridi’s argument is not that machines have become human, but that humans can no longer define themselves only by reference to information processing.

His language of the “infosphere” matters here. Human beings increasingly live in an onlife condition in which online and offline experience blend, and in which information-processing systems silently shape attention, choice, visibility, and memory.

Floridi’s response is not despair but recalibration. Intelligence, on this account, is not exhausted by computation; it includes embodiment, temporality, mortality, care, vulnerability, and the perspective that comes from being a finite creature with a history.

What Remains

The question “what does it mean to be human” is not answered by AI. It is sharpened by it. Systems that mimic cognition without possessing lived experience return us, more urgently than before, to the features of human life that resist that mimicry.

Those features are embodied, contextual, temporal, mortal, and fundamentally relational. They are not optional add-ons to intelligence. They are the conditions under which meaning, care, accountability, and genuine understanding become possible at all.

As of March 2, 2026, debates over AI safety, alignment, education, labour, and emotional dependence were already circling these human-centred questions. The risk is not only that AI might become more powerful, but that in optimising for prediction, convenience, and engagement, societies may neglect the slower forms of attention and mutual recognition through which people become selves.

At the threshold between the world AI described and the world AI is creating, that clarity is not a luxury. It is the foundation on which everything that follows will have to be built.

Reference List

Floridi, L. (2014). The Fourth Revolution: How the Infosphere Is Reshaping Human Reality. Oxford University Press.

Searle, J. R. (1980). Minds, brains, and programs. Behavioral and Brain Sciences, 3(3), 417–424. https://doi.org/10.1017/S0140525X00005756

Turing, A. M. (1950). Computing machinery and intelligence. Mind, 59(236), 433–460. https://doi.org/10.1093/mind/LIX.236.433

Turkle, S. (2011). Alone Together: Why We Expect More from Technology and Less from Each Other. Basic Books.

LLM Timeline — Frontier AI Model Release Tracker. Accessed for contextual framing dated to March 2, 2025. https://llm-release-dashboard.vercel.app

2025 AI Timeline. Hugging Face Space. Accessed for contextual framing dated to March 2, 2025. https://huggingface.co/spaces/2025-ai-timeline/2025-ai-timeline

Author Note

AI Usage: This article was drafted with assistance from a generative AI system to organise structure and suggest phrasing. All facts, citations, and final editing have been verified and approved by the author. The AI did not access any private health data.