PARIS: Major AI company executives are spreading the myth that “strong” machine intelligence will soon surpass human intelligence, but many experts in the field dismiss these promises as marketing gimmicks.
The idea that current machine-learning techniques would produce “artificial general intelligence” (AGI), which is often referred to as human-or-better intelligence, feeds theories about the future that range from human extinction to machine-delivered superabundance.
“Systems that start to point to AGI are coming into view,” said Sam Altman, the head of OpenAI, in a blog entry last month. The milestone “may come as early as 2026,” according to Dario Amodei of Anthropic.
These forecasts serve as justification for the hundreds of billions of dollars being spent on computer technology and energy.
