Elon Musk predicts that artificial intelligence will exceed humanity’s combined intelligence within roughly five years.
He warns that people will likely lose control of AI within the next ten years, even as the technology generates unprecedented abundance.
Musk urges leading AI developers to conduct regular joint reviews of frontier models to assess safety risks before release.
Musk estimates that AI will exceed humanity’s combined intelligence in roughly five years, and that people will lose control of the technology within a decade.
During a conversation with The Economist’s editor-in-chief Zanny Minton Beddoes on Thursday, Musk, founder of xAI, said AI is on track to outperform humans in virtually all intellectual pursuits.
Musk stated, “AI will be capable of doing anything better than humans, except for being human itself. The most probable scenario is an era of extraordinary abundance where anyone can obtain anything they imagine. Though this may seem fantastical, we are now in 2026; let’s assess our position by 2036.”
This conversation marks the latest occasion on which Musk asserted that AI will usher in an age of abundance, delivering both benefits and existential concerns. Nonetheless, he maintains that advanced AI carries significant risks and that there is currently no viable means to decelerate its development.
Musk added, “I cannot envision any effective way to halt this unstoppable momentum of AI and robotics. In fact, even if a stop button existed, I doubt we should press it, given the likely outcome of universal abundance.”
He advocated for regular meetings among the world’s premier AI companies to discuss safety and security issues, and proposed granting competing labs limited early access to each other’s frontier models so they can detect hazardous capabilities before release. He warned that companies failing to mitigate serious risks could prompt governmental intervention.
Although he urged cooperation among top AI developers, Musk also criticized his longtime rival Sam Altman and OpenAI’s shift from a nonprofit to a for‑profit entity, arguing that this departure undermines its original mission.
He said, “I am not a fan of Sam Altman, because you launched a nonprofit intended to be an open‑source, globally owned AI company, yet it has become an $800 billion for‑profit venture with closed‑source technology.”
In May, a California jury dismissed Musk’s $150 billion lawsuit against OpenAI, CEO Sam Altman, and co‑founder Greg Brockman, ruling that the defendants were not liable for abandoning the organization’s nonprofit mission by adopting a commercial structure. The decision concluded one of Musk’s most prominent legal battles with the ChatGPT developer.
Musk suggested that Anthropic was founded because co‑founder Dario Amodei and his team no longer trusted OpenAI’s CEO Sam Altman, contending they would otherwise have stayed at OpenAI.
He noted, “I consider Dario a very principled individual who cares about the world’s future, and everyone I have met at Anthropic so far has been well‑intentioned. No one has triggered my ‘evil detector.’ The path to disaster is largely paved with bad intentions; even a few well‑intended stones remind us not to become complacent.”
Although Musk commended Amodei’s leadership at Anthropic, critics such as Sam Altman have accused the company of employing fear‑based marketing to promote its Claude AI products.
Even with his disagreements with OpenAI’s leadership, Musk asserted that rival AI companies ought to cooperate on safety matters.
He concluded, “In the end, if we need to converse, we will; let us set aside personal differences for the greater good of the world.”
