The technological singularity is the hypothetical moment when machine intelligence surpasses human intelligence and the future becomes impossible to foresee from here, like the inside of a black hole for an outside observer: hence the metaphor borrowed from physics. The earliest known mention is Stanisław Ulam's: in 1958 he recalled a conversation with John von Neumann about technical progress that seemed to approach "some essential singularity" in the history of the race (Ulam, 1958). The name was fixed by the mathematician and novelist Vernor Vinge in 1993: "Within thirty years, we will have the technological means to create superhuman intelligence" (Vinge, 1993). Ray Kurzweil gave it a date (2045) and turned it into a program of exponential optimism (M. Mitchell, 2019, ch. 3).
The conceptual engine is Good's "intelligence explosion": an ultraintelligent machine could design better machines, and so on (Good, 1966, p. 33) (see FOOM and recursive self-improvement). In the song it arrives unannounced, after an uneventful training run: "But now the singularity's begun" (osmarks, n.d.).
Today the term sounds somewhat dated in San Francisco, where people prefer to speak of AGI, "takeoff" or superintelligence; recall that MIRI was called the Singularity Institute until 2013. Criticism comes from two sides. One is the faith in exponential curves: Mitchell reports that Andrew Ng rolled his eyes at the mention of Kurzweil, and quotes the journalist Andrian Kreye, who called the idea "nothing more than the belief in a technological Rapture" (M. Mitchell, 2019, ch. 3). The other is its eschatological structure: a date, a threshold, a transfigured humanity. That the singularity so closely resembles Teilhard's Omega Point is no accident, and Santelli (2026) notes that vague prophecies have disappointed humanity too often to trust one more messianic movement.