Elon Musk says his AI safety concerns date back more than a decade

“I was thinking about AI safety long before 2014,” he said on US social media platform X, which is part of his artificial intelligence venture SpaceXAI.

Musk pointed to his involvement with philosopher Nick Bostrom's 2014 book Superintelligence: Paths, Dangers, Strategies, saying he contributed to the work and was thanked by name in its foreword.

The book examines the potential consequences of AI surpassing human intelligence, warning that humanity's survival could hinge on engineering precise control mechanisms before superintelligent systems emerge.

Musk also pointed to a post from August 2014 in which he recommended the book and warned that advanced AI was “potentially more dangerous than nukes.”

His comments come as senior technology executives call for slowing the development of cutting-edge AI systems over safety concerns.

Over the weekend, Anthropic Chief Executive Dario Amodei proposed giving independent safety evaluators access to AI companies to verify their safety practices, an approach backed by OpenAI Chief Executive Sam Altman and Musk.

Amodei told CBS Sunday that “there are real dangers," about the development of AI.

"If we build in the right way, I think the probability of something bad happening is very low," he said. “If we build in the wrong way, the probability of something bad happening is very high.”