Elon Musk has issued a stark warning about the future of artificial intelligence, arguing that AI’s rapidly growing energy demands may soon outpace Earth’s ability to generate power. Speaking on a recent podcast, Musk said discussions around AI often focus on software and innovation while ignoring a more basic constraint: electricity. Massive data centers, specialized chips, and cooling systems already consume enormous amounts of energy—and that demand is accelerating fast.

Musk noted that the entire United States currently uses roughly half a terawatt of electricity on average, and doubling that output would require decades of infrastructure expansion. Power plants, transmission lines, and grid upgrades simply don’t scale at the same speed as AI development. In his view, this mismatch could become critical within the next few years if current trends continue.

Looking ahead, Musk suggested that space may offer a solution. He predicted that within the next 30 to 36 months, hosting large-scale AI infrastructure in orbit could become more practical and cost-effective than relying solely on Earth-based systems. In space, AI facilities could take advantage of near-constant solar energy, avoid land and cooling constraints, and reduce the need for large battery storage.

As the founder of SpaceX, Musk has already hinted at plans that align with this vision, including large satellite deployments that could support orbital, solar-powered infrastructure. While space-based AI remains speculative, his comments reflect a growing concern across the tech and energy sectors: if artificial intelligence continues to expand at its current pace, the real bottleneck may not be computing power—but how much energy humanity can generate and sustain.