In a recent interview with the editor-in-chief of The Economist, Musk stated that artificial intelligence could surpass the combined intelligence of all humanity within about five years, and that humans may not be able to maintain control of this technology for the next decade. At the same time, he believes the trend of AI and robotic expansion is difficult to reverse.
Musk predicts that AI will surpass humans in every way.
Musk claims that AI will surpass humans in almost all intellectual tasks in the future, with the only exception being "humanity itself." He believes the more likely outcome is not a complete shutdown, but a significant increase in productivity, leading to unprecedented material abundance.
He stated that he sees no real way to stop the continued advancement of AI and robotics. Even if a "stop button" exists, it shouldn't necessarily be pressed, because in his view, this technology is more likely to bring about widespread resource abundance than one-way destruction.
Propose that leading companies conduct cross-inspection of cutting-edge models
On security issues, Musk suggested that major global AI companies should hold regular meetings to discuss model security and potential risks. He also recommended that competitors share their cutting-edge models with each other to a limited extent before publicly releasing them, in order to identify potentially dangerous capabilities in advance.
According to his vision, the government should only intervene if a company fails to address a serious risk after it has been identified. This suggests that he prefers to first establish a collaborative mechanism within the industry before introducing regulatory power.
OpenAI criticized again for deviating from its original positioning.
When discussing OpenAI, Musk again singled out CEO Sam Altman for criticism, stating that he did not agree with OpenAI's path from a non-profit organization to a high-valuation commercial company, and believed that this deviated from its original positioning of "facing the world and emphasizing open source".
In May of this year, a California jury dismissed Elon Musk's $150 billion lawsuit against OpenAI, Altman, and co-founder Greg Brockman. The case alleged that the parties deviated from their non-profit mission during the organization's commercialization process, but the jury found the defendants not liable, thus ending one of Musk's most high-profile legal challenges against OpenAI.
Musk also mentioned Anthropic, saying that if Dario Amodei's team still trusted OpenAI's management, Anthropic might not exist in its current form. However, he also affirmed Amodei's personal stance, stating that he values the impact of AI on the future of the world.
Despite his long-standing disagreements with OpenAI's management, Musk stated that major AI companies should be willing to cooperate on security issues and set aside personal differences if necessary.
