Elon Musk warns China will lead global AI power race

Elon Musk warns China leads AI

On the Moonshots with Peter Diamandis podcast, Elon Musk said he believes China could surpass other countries in aggregate AI computing power if its energy and infrastructure plans come to fruition.

While the US focuses on restricting advanced semiconductor access, Musk argued that China will eventually “figure out the chips.” The combination of electricity capacity and scalable infrastructure could allow China to outpace other nations, signaling a shift in AI dominance that governments and companies need to watch closely.

China’s electricity boosts AI power

Musk highlighted that electricity generation is the real bottleneck for scaling AI systems. He projected on the podcast that by 2026, China could produce up to three times the electricity output of the United States and said that such capacity would materially ease the power constraints facing energy-intensive AI data centers.

The emphasis on energy also explains why AI centers require power levels similar to those of small cities. Musk warned that many are underestimating how difficult it is to bring electricity online quickly. Countries that fail to expand their power infrastructure may lag in AI capabilities despite having advanced chips or software talent.

Power limits next AI breakthroughs

Several industry voices, including Musk and some analysts, argue that energy supply may become a primary constraint on scaling AI, even as chips and algorithms remain important.

Analysts at Goldman Sachs have warned that electricity and grid constraints could slow AI expansion in the United States and have pointed to China’s projected spare generation as a possible advantage.

China may outpace US in AI

Musk suggested China could surpass the United States in aggregate AI compute and pointed to slowing incremental gains at the cutting edge of chip performance as one reason that energy and scale could become decisive factors.

Even without access to the most advanced semiconductor designs, China’s combination of energy surplus and manufacturing scale may allow it to catch up quickly.

This perspective challenges the assumption that limiting chip exports is enough to contain China. Musk’s warning suggests a broader strategic approach may be needed, focusing on infrastructure, energy, and research to remain competitive in the global AI landscape.

China’s chip strategy evolving fast

Despite export restrictions, Musk believes China will develop its own chip capabilities. He noted that slowing advances at the highest performance levels may give China time to innovate independently, potentially closing gaps in AI hardware more quickly than many experts anticipate.

China’s focus on homegrown chips aligns with its broader technology ambitions, creating a feedback loop where energy capacity and semiconductor development reinforce each other. Musk’s comments highlight that AI dominance depends on more than software or algorithms; hardware and power infrastructure are equally critical.

AI demands city-scale electricity

Training advanced AI models consumes enormous power, sometimes comparable to what a small city uses. Musk warned that scaling AI without reliable electricity is nearly impossible, making energy supply a strategic factor in the global AI race. This advantage may allow China to operate larger, faster AI systems than its competitors.

The implication is clear: countries that fail to upgrade grids and energy infrastructure could struggle to support the next wave of AI breakthroughs. Musk’s comments draw attention to the often-overlooked link between power generation and technological leadership.

US faces AI power bottlenecks

Analysts from Goldman Sachs have highlighted that the US could face slowdowns in AI progress due to electricity limitations. Musk echoed these concerns, emphasizing that scaling AI requires infrastructure capable of supporting extremely high energy demand for training large models.

The warning is significant as many US companies focus on hardware and software innovation, but may underestimate the grid constraints. Without addressing energy supply, efforts to maintain leadership in AI could face serious hurdles compared to nations with abundant power.

China’s spare energy aids growth

Goldman Sachs projects that by 2030, China could have roughly 400 gigawatts of spare generation capacity, a level analysts say would be large relative to projected global data center demand.

Musk pointed out that this surplus positions China to maintain AI growth without the same bottlenecks faced by other countries, giving it a clear competitive advantage in the AI sector.

Musk’s insights suggest that infrastructure, not just talent or technology, will determine which country dominates AI in the coming decade.

China integrates AI with industries

Chinese leader Xi Jinping highlighted advances in AI and local chip development in public remarks at the end of 2025 and in his New Year address, which media outlets reported as framing 2025 as a year of breakthroughs.

The country’s focus on merging AI with economic sectors demonstrates how infrastructure, policy, and technology converge to accelerate innovation. This coordinated strategy could further widen the gap between China and other nations in AI capabilities in the years ahead.

Global AI innovation priorities

Musk’s commentary also serves as a warning for other nations: AI leadership is increasingly tied to power infrastructure and industrial coordination. Countries focusing solely on software or algorithms may fall behind if they neglect energy capacity and large-scale AI deployment capabilities.

This shift suggests that global AI strategies must include investments in electricity, manufacturing, and research ecosystems. Musk’s insights show that multi-layered planning is crucial to maintain competitiveness in the rapidly evolving AI landscape.

AI’s next bottleneck is power

Musk stressed that future AI breakthroughs depend on solving energy bottlenecks. Even the most advanced algorithms require data centers that can operate continuously at high power. Without adequate electricity, AI development could stall despite innovations in software or hardware design.

This view highlights that countries with robust energy planning will likely dominate AI. Musk’s focus on electricity underscores a shift in what matters most for AI supremacy: infrastructure and power, not just chips or models.

Diminishing returns aid China’s catch-up

Musk noted that cutting-edge chip performance is hitting diminishing returns, which may allow China to catch up more easily in hardware. Even without top-tier semiconductors, its energy advantage and production scale could accelerate its AI development faster than expected.

This factor, combined with growing infrastructure and coordinated industrial policy, could give China a unique path to AI dominance. Musk’s warning suggests the US and others cannot rely solely on export restrictions to maintain an edge.

Strategic AI lessons for nations

The global AI race now depends on multiple factors: energy infrastructure, chip development, and industrial coordination. Musk’s insights highlight that countries planning holistically are more likely to maintain leadership in AI over the coming decade.

Investments in electricity grids, domestic chips, and large-scale AI centers are no longer optional. Nations ignoring these elements risk falling behind as China leverages its advantages to lead the next wave of AI innovation.

Can vibe coding really replace a computer science degree? See why this CEO says tech careers are changing for good.

Conclusion on AI global race

Elon Musk’s warning about China underscores how AI supremacy is shaped by more than algorithms. Energy, chips, and industrial coordination play crucial roles, making infrastructure a central factor in who leads the next generation of AI innovations.

Nations ignoring these aspects may fall behind quickly.

Want my must-have Chrome extensions? Check out these 11 Chrome extensions that go on all my new PCs.

What do you think about the global AI power shift? Share your thoughts.

This slideshow was made with AI assistance and human editing.

Don’t forget to follow us for more exclusive content on MSN.

Read More From This Brand:

• Samsung plans to start US 2nm chip manufacturing by 2026
• Break these bad tech habits before 2026
• 2026 set to be AI monetization year, says Wedbush analyst