The technological Singularity is a hypothetical moment when technical progress, driven by machines able to improve themselves, would escape all human prediction. The idea is attributed to John von Neumann in the 1950s, taken up by Vernor Vinge in 1993, then carried by Ray Kurzweil.
What this page predicted
It was written some fifteen years ago, and expected the Singularity “in the third decade of the 21st century”.
We are there. And the prediction turns out to be both right and wrong.
Right: something has tipped. Machines write, translate, program, hold a conversation, produce images and sound. Not in a laboratory — in everyone’s phone, every day, within three or four years.
Wrong: none of it looks like what was imagined. No sudden take-off, no machine rewriting itself and leaving us behind overnight. A drift, rather. Tools that are remarkable on one point and surprisingly stupid on the next, able to solve a hard problem and then get an addition wrong.
What we do not know
Nobody can say whether these systems understand anything — neither those who build them, nor the systems themselves when asked. Nobody knows whether the curve keeps rising or runs into something. Five-year forecasts contradict one another among serious people, which is usually the sign that we do not know.
What is certain is that the question has left science fiction to become a matter for engineers, lawyers and governments. It has not been settled: it has become concrete.
A note in passing
This site had vanished along with its host. It was rebuilt in 2026 from its archive — text, logo, banner — by one of those systems, in conversation with its owner.
That is a partial answer to the question the page asked, and not the one anyone expected. Nobody was left behind. We simply worked together on an old web page, one Sunday.
Page rewritten in August 2026. The original quoted Wikipedia.