Who Coined the Term 'AGI'? The Real Inventor Finally Identified

Artificial General Intelligence, or AGI, is one of the most prominent moving goalposts in the tech community right now. Each breakthrough in artificial intelligence (AI) research seems to redraw the boundary of what “general” intelligence means. For scientists, it's a long-term research aspiration towards a machine that is capable of true innovation; for companies, it is the ideal assistant that can perform all the corporate tasks a human can do; and for the layman, it is the manifestation of doomsday.

In that sense, AGI has become a kind of Rorschach test, reflecting the hopes, fears, and motivations of whoever invokes it. Yet before it turned into a catch-all for the industry's future visions, the idea had a specific origin, and one individual was responsible for naming it.

The Popularisation of AGI

The AGI rabbit hole is pretty deep. In recent years, pretty much every company's town hall in Silicon Valley consists of this acronym, with every speaker defining it loosely and connecting it to their vision. But things are a bit different in the scientific community, where the topic has been a subject of research for decades.

A quick search on Google will give you the names of Ben Goertzel and Cassio Pennachin, the co-editors of a book titled Artificial General Intelligence. They came up with the phrase while finding a replacement name for the working title of the book “Real AI”. The book was published in 2005, and this is largely the point of the popularisation of the term.

In the book, they defined AGI as (via Forbes) “AI systems that possess a reasonable degree of self-understanding and autonomous self-control, and have the ability to solve a variety of complex problems in a variety of contexts, and to learn to solve new problems that they didn't know about at the time of their creation.”

Goertzel's usage of the phrase is considered the origin point because just a couple of years later, the first Conference on Artificial General Intelligence was held in 2008. The conference was organised by the AGI society and was led by Goertzel and Marcus Hutter. It has now become a global annual event, which has attracted speakers such as the Turing Award winner and one of the three Godfathers of AI, Yoshua Bengio.

However, Goertzel himself acknowledges that he was not the inventor of the phrase.

Who Really Coined the Term AGI?

In 2011, Goertzel penned a blog post titled “Who coined the term AGI?” In the post, he acknowledged the shortcomings with the phrase and admitted that he was, in fact, not the first one to use the phrase.

He credits Mark Gubrud, a Maryland, US-based researcher, who came up with the term in 1997 in a paper titled “Nanotechnology and International Security.” According to a Wired report, the paper was published and presented at the Fifth Foresight Conference on Molecular Nanotechnology, where he used the term to describe how future technology can revolutionise US military affairs.

He also defined the technology as “AI systems that rival or surpass the human brain in complexity and speed, that can acquire, manipulate and reason with general knowledge, and that are usable in essentially any phase of industrial or military operations where a human intelligence would otherwise be needed.”

While these definitions are dated, the commonality between these and the one given by OpenAI CEO Sam Altman is that it is an AI system that operates at a level similar to humans without any supervision. With Google, OpenAI, Microsoft, Meta, and countless other AI companies racing to reach this target as soon as possible, it is interesting to reflect upon the fact that the concept emerged from a little-known researcher who believed in 1997 that technology had the power to replace the human mind.