What “the singularity” means, and why AI researchers still disagree on 2047
GregCook2011 · x · 2026-07-26
What “the singularity” can mean
The post disentangles three different meanings of “singularity”:
- Vernor Vinge (1993): the point where machines outthink humans and forecasting breaks down.
- I. J. Good (1965): a recursive machine-that-builds-better-machines loop.
- Ray Kurzweil: a smoother exponential trajectory, with arrival around 2045.
It argues that science has no single consensus on the idea because it was never a standard research program with a clear measurement. The best available evidence comes from surveys and economics: a large survey of publishing AI researchers puts human-level machine intelligence around 2047, but with enormous disagreement, and a majority doubt that scaling current methods is enough to get there.
The quoted context also notes Sam Altman’s “we are now in the singularity” claim, contrasting it with the fact that most people outside the AI bubble still don’t grasp what is happening.
More from AGI Musings
- A reply repeats Altman’s claim that superintelligence is on a clear glide path — haider1 · 2026-07-26
- Sam Altman says superintelligence is on a clear glide path and will get built — haider1 · 2026-07-26
- Employee Automates Entire Job with AI, Gets Promoted to Head of AI — tomcrawshaw01 · 2026-07-26
- Universal high income, the author says, only becomes viable once permanent unemployment is obvious — rand_longevity · 2026-07-26
- AI Bubble Risk Could Be Worse Than the Dot-Com Bust Because So Much More of It Is Debt-Funded — rohanpaul_ai · 2026-07-26
- From Mockery to Business Leverage: The Decade Evolution of Musk's AI Safety Rhetoric — theteknosaur · 2026-07-26