It's not just a technological revolution! Elon Musk predicts: In 2026, a major reshuffle for humanity—these types of people may be replaced first.

Elon Musk has consistently pushed forward with the plan to reuse rocket components in space transportation projects. In multiple missions carried out by SpaceX, the boosters successfully return and land after completing their work, which has significantly increased the frequency of space launches. Operations that once seemed difficult to achieve, like recovery, have now become routine, with the team recording data after each mission and optimizing for the next. On the automotive side, Tesla’s electric vehicles enter the market through factory assembly lines, gradually replacing some traditional powertrain vehicles in practical use. Advances in production and testing have made these models perform more stably on the road. Brain-machine interface technology is also moving from the lab to application, with Neuralink’s project allowing some test participants to control devices directly through brain signals, completing simple operations and interactions—these progressions are reflected in real-world records.

In a recent multi-hour podcast discussion, Musk shared his views on technological development over the next decade. The conversation took place in early 2026, and he provided a clear timeline based on current computing power and industry conditions. He mentioned that general artificial intelligence could be achieved by 2026, and by 2030, the total intelligence of AI might surpass that of all humanity combined. These judgments are not casual guesses but are based on observed accelerating technological trends. During the discussion, he repeatedly emphasized that AI development has reached a critical point; some predictions from the past decade are gradually materializing, and now the focus should be on the overall structural changes ahead.

He specifically analyzed how different types of jobs will be affected. He pointed out that roles primarily relying on screens and keyboards to process information might face early adjustment pressures because these jobs essentially involve inputting data, organizing it into conclusions, and outputting reports or charts—areas where AI shows clear advantages in speed and continuity. In contrast, tasks requiring direct physical interaction with the world, such as caring for children or on-site pipeline installation, will experience different pacing adjustments. He did not shy away from reality but broke down the essence of work to clarify, avoiding the misconception that change is sudden or unexpected.

Power supply was identified as a fundamental constraint in AI development. He explained that large data centers require enormous amounts of electricity to run chips, and also need solutions for heat dissipation and grid stability. While chip production is rapidly increasing, infrastructure development is progressing more gradually, and this speed gap will become prominent around 2026. The process of connecting data centers to power grids itself takes time; he mentioned that his company’s supercomputing clusters took a long time to coordinate with the grid, even in regions with relatively mature infrastructure. Once power cannot keep up, even the most advanced chips cannot perform effectively, so those who have an advantage in energy supply will be able to sustain AI operations longer.

Regarding regional differences, he directly compared the expansion of electricity capacity in various places. China’s growth in power generation capacity, especially through solar and other renewable energy deployments, is rapid. He estimates that by 2026, this power gap will give China a clear advantage in supporting large-scale AI computations. The U.S. grid largely relies on infrastructure from the last century, with long upgrade cycles, whereas China’s ultra-high-voltage technology and new energy grid integration are leading globally. These observations are based on publicly available energy development data, not mere speculation. He emphasized that electricity will become a true strategic resource, determining how long AI can run and how many tasks it can handle simultaneously, rather than just the number of chips.

As the discussion circulated, attention turned to the reshaping of employment structures. He mentioned that the next three to seven years will be a period of significant adjustment, with traditional jobs changing due to automation, requiring many people to redefine their roles in the workforce. Production costs will increasingly focus on raw materials and energy, making many goods more accessible. Society will experience an abundance of items, but as employment patterns are disrupted, new rhythms will need to be adopted. Just as certain industries adjusted due to technological proliferation in the past, this change will be broader in scope, with the core goal of helping people understand the actual trends.