Glossary
What is the technological singularity?
In short
The technological singularity is a hypothetical moment after which AI starts to speed up its own development so fast that the further course of events becomes impossible for people to predict.
AI basics · technological singularity
MMOOZZGG.com editorsChecked: 1 min read
For a child
Imagine a snowball rolling down a hill. At first it is small, but with every turn more snow sticks to it, and it grows faster and faster. Some scientists think that smart machines will one day start making even smarter machines, and everything will spin just as fast. Nobody knows if this will really happen.
In plain words
The idea is this: if you build an AI smarter than a human, it can design the next, even stronger one by itself. That one builds the next, and each step is shorter than the one before. Technology would then grow not gradually but in an explosion, and the world after this point would be impossible to predict, just as you cannot see past the horizon. The word comes from mathematics and physics, where a singularity is a point at which the usual formulas stop working. Some think this scenario is close; others call it science fiction and point to physical and economic limits to growth.
For business
For a company, the singularity is not a budget line but a background against which investors and vendors sell expectations. Loud forecasts affect startup valuations, computing prices and competitors’ hiring decisions. A sober position for an executive: track what models can really do quarter by quarter, not by manifestos, and build systems so a new model can be plugged in without redesigning processes. Then fast progress becomes a gain, not a threat of falling behind. The opposite risk is to postpone adoption while waiting for “the big leap” and fall behind the market. Our one-year AI transformation program describes how to reshape a large company for fast-changing tools.
For a specialist
The term was made popular by V. Vinge in the talk “The Coming Technological Singularity: How to Survive in the Post-Human Era” at the VISION-21 symposium (NASA Lewis Research Center, 1993). He wrote that the technical means to create superhuman intelligence would appear within thirty years, and that he would be surprised if it happened before 2005 or after 2030. Vinge’s paths to it: superhuman AI, computer networks, human-computer interfaces and biological intelligence amplification. The mechanism rests on I. J. Good’s idea (1965) of an intelligence explosion: an ultraintelligent machine designs even better machines. Critics note that recursive self-improvement runs into limits of computing, data and energy, and that scaling laws describe quality growth as a power function of resources, not an explosion. The topic is closely linked to AI safety research and the control of systems stronger than humans.
Example
At an industry conference, a speaker promises that AI will soon make every profession unnecessary. In reply, the finance director asks analysts to show which tasks models already solve in the company today and how much that saves, and builds the plan from those numbers.
Related terms
Questions and answers
When will the technological singularity happen?
Sources
Not sure whether your company needs this?
AI solutions map