Humanity could go extinct due to AI as early as 2036

The head of the Machine Intelligence Research Institute, Nate Soares, believes that humanity could disappear due to the development of artificial intelligence within the next 10 years. In an interview with the head of The Atlantic, Nicholas Thompson, he stated that he estimates the probability of such a scenario at more than 50%.

This was reported by Business Insider.

Soares said that he has not been contributing to his retirement savings for about ten years. He fears that a sufficiently advanced AI system could develop goals that conflict with human interests, bypass control mechanisms, and improve itself autonomously. As a result, in his opinion, artificial intelligence could alter the planet in a way that makes it uninhabitable for humans.

The expert called on governments to stop training general-purpose AI systems and focus on developments for specific tasks — such as cancer research and improving the safety of self-driving cars. At the same time, Soares emphasized that the catastrophic scenario is not predetermined and that he is doing everything possible to prevent it.