© Skydance Media
Ten years, five years, even two years. This is how much time humanity would have left before an apocalypse caused by machines and artificial intelligence. This dark scenario, which is of course far from unanimous, is proposed by Eliezer Yudkowsky, a 44-year-old academic, who has worked in the past on the development of AI.
A scenario between Matrix and Terminator
Today, the latter is a principal investigator of a non-profit organization called Machine Intelligence Research Institute in Berkeley, California. In a fascinating article, our colleagues from the Guardian set out to meet these techno-pessimists.
In Eliezer Yudkowsky's idea, we get closer to what we have already seen at the cinema in popular films such as Terminator or Matrix. And the expert says: “The problem is that people don't realize it. We have a tiny chance that humanity will survive.”
Why ? According to him, AIs are evolving very quickly and they will soon turn on their designers. He does not imagine a machine designed like a super-brain but rather “a kind of extraterrestrial civilization that thinks a thousand times faster than us” scattered in a large number of boxes which we could no longer dismantle.
In this scenario worthy of the best nuggets of science fiction, humanity could only escape through targeted air strikes, or even outright use of nuclear weapons. In any case, we suggest you read the numerous testimonies cited by the British media which are worth the detour. Some of them are much less radical and express fairly legitimate concerns about recent technological developments.
Real fears about AI
Eliezer Yudkowsky is far from being the only specialist worried about the future of AI. We recently spoke to you about this letter sent by several researchers to the board of directors of OpenAI, the company that develops ChatGPT.
In this letter, experts speak of a discovery so powerful in artificial intelligence that it could threaten humanity. Unfortunately, we were unable to review this document. But it could be a reference to artificial general intelligence (AGI).
If you're not really familiar with this notion, know that it is a question of a system which would become autonomous and could therefore function, and above all reason, without human intervention. We would therefore echo the fears of an AI that could make its own decisions without supervision. To learn more about this, you can always re-read our dedicated article here.
📍 To not miss any news from Presse-citron, follow us on Google News and WhatsApp.
[ ]