Why AI Cannot Know What It Feels Like to Be Alive
Three hundred billion miles from infinity.
Iain McGilchrist argues that AI can accumulate facts but never achieve genuine understanding, because real knowing requires embodied, mortal interiority — a felt inside that no amount of data or processing power can simulate or substitute.
Actions
The Source
The Counterintuitive Need to Slow Down and Find Spaciousness with Iain McGilchrist | TGS 217
The Observer
Hemisphere theory, neuroscience, philosophy of mind — left and right brain as modes of being, the crisis of left-hemisphere dominance, and the nature of consciousness
The Translation
AI-assisted summaryFamiliar terms
Iain McGilchrist identifies a philosophical catastrophe hiding in plain sight within the English language: the collapse of two fundamentally distinct modes of knowing into a single word. French maintains the distinction between 'savoir' — propositional, factual knowledge — and 'connaître' — knowledge born of direct, embodied acquaintance. This linguistic poverty, McGilchrist contends, has made it dangerously easy for anglophone culture to mistake information processing for genuine cognition, and to treat AI-generated outputs as though they constitute understanding.
The argument is categorical, not quantitative. AI can possess 'savoir' in extraordinary abundance — it can retrieve, correlate, and recombine factual content at scales no human mind could match. But 'connaître' requires something AI structurally lacks: interiority. To know grief, joy, mortality, or embodiment from the inside demands a phenomenological subject — a being for whom experience has qualitative character. McGilchrist deploys a striking mathematical analogy to drive the point home: extending a fraction of an inch to 300 billion miles brings you no nearer to infinity. The distance between AI and genuine understanding is not a technical gap awaiting closure through increased parameters or training data; it is an ontological chasm.
This framing challenges the implicit materialism underwriting much AI discourse, which assumes that cognition is substrate-independent computation. McGilchrist insists that emotions, embodiment, and mortality are not incidental accompaniments to thought but constitutive conditions of deep understanding. Without an inside — without felt, mortal interiority — there is no understanding to speak of, only its increasingly convincing simulation.