Elon Musk, the founder of Tesla and SpaceX, said in a recent post on X that artificial intelligence could exceed the combined intelligence of all humans within four to five years. He described the coming wave as a "super sonic tsunami" that would sweep the world, underscoring his increasingly aggressive timeline for AI progress.
The article notes that Musk has repeatedly shortened his forecasts. At the World Economic Forum in Davos earlier in 2026, he said AI could be smarter than any individual by the end of 2026 and smarter than all humanity by the start of the next decade. Earlier this year, at the Abundance conference, he said artificial general intelligence, or AGI, could arrive in 2026. In April 2024, he had said systems smarter than any person might appear by 2025 or 2026, and later that year he suggested AI could match the output of eight billion people within a few years of reaching human-level performance.
Musk is not alone in predicting rapid gains. Anthropic CEO Dario Amodei estimated that by 2027, systems could emerge with intellectual output comparable to "a whole country of geniuses." Industry figures also point to sharply falling costs, saying model usage prices have dropped by hundreds of percent over the past two years, and to the possibility of recursive self-improvement, where AI systems create more advanced versions of themselves.
The piece also highlights skepticism. Critics say Musk’s past timelines have often not materialized on schedule, and others argue that intelligence is more than computation or information retrieval, with judgment, morality, and deep human understanding still much harder to replicate. There is also concern that such powerful tools could create unprecedented risks if they fall into the wrong hands, even as the AI race continues at speed and the next few years could prove decisive for the relationship between humans and machines.
