Elon Musk on AI: humans will no longer be in control in ten years | The Economist

Artificial intelligence is projected to exceed the collective sum of human intelligence by approximately 2029, potentially ushering in an era of unprecedented material abundance. Elon Musk, founder of xAI and co-founder of OpenAI, suggests that while a 10 to 20 percent risk of existential catastrophe exists, the momentum of AI development is now inexorable, necessitating a shift toward optimistic engagement rather than futile attempts at cessation. To mitigate safety risks, leading AI firms should establish a self-policing framework where competitors grant each other early access to new models to identify security flaws before public release. This peer-review system would leverage competitive incentives to ensure transparency, involving government intervention only when a company refuses to address identified hazards. Despite personal friction among industry leaders, such as the ideological divide regarding OpenAI’s transition to a closed-source, for-profit entity, collaborative safety protocols remain essential for the global interest.