Xi-Trump Pause on Rare-Earths Tariff Shields AI Bubble Temporarily

On 10 October, a remark made by China’s former leader Deng Xiaoping was recalled around the world, becoming a focal point of commentary across Europe.

Deng had stated in 1992, during his southern tour that ‘China’s rare-earth resources account for 80 per cent of the world’s known reserves. Their importance can be compared to Middle Eastern oil, possessing extremely vital strategic significance. We must handle rare earths well and make full use of China’s advantages in this field.’

Whether Deng Xiaoping actually uttered these exact words remains unverified. Nevertheless, his emphasis on rare earths and his far-sighted vision for China’s resource strategy equal, if not surpass, the perseverance celebrated in Confucian culture: the enduring patience of ‘sleeping on brushwood and tasting gall’, undertaken to engrave in memory past lessons, and imbue the resolve to rise again.

During the 1990s, Chinese engineers mastered heavy rare-earth separation technology and secured China’s near monopoly in production. Deng’s trusted geologist at the time, Wen Jiabao, who later served as Premier of China from 2003 to 2013, claimed to have been directly involved in nearly all rare-earth policies.

In Deng’s era there were no smartphones and artificial intelligence existed only in science-fiction films. Yet China already understood the strategic value of rare earths. By 2025, their importance had become irreplaceable. Rare-earth metals are known as ‘the vitamins of modern technology’. Without them, there would be no iPhone, no precision-guided missiles and no AI.

Today the Middle East holds roughly half of the world’s oil reserves, while China dominates about 85 per cent of global refining and 90 per cent of rare-earth permanent magnets production. In a sense, Deng Xiaoping’s 1992 statement did not fully capture the future potential of rare earths.

It is safe to say that rare earths now underpin 21st-century technological competition and defence systems. That explains why, when China’s Ministry of Commerce announced its ‘decision to implement controls on exports of rare-earth-related items to foreign countries’ on 9 October, Donald Trump reacted sharply and claimed he saw no reason to proceed with the Xi-Trump summit. Markets, and indeed the world, were instantly choked by renewed fears of a trade war.

‘I wonder if that timing was coincidental?’ Trump wrote on Truth Social, implying that China’s announcement was ‘especially inappropriate’ as it undercut his Middle East peace effort.

Just a Bad Moment?

9 October was the first working day after China’s National Day holiday, when the government announced export controls at exactly 9 a.m., possibly not by coincidence.

While there was no indication that the decision was aimed personally at President Trump or linked to the Middle East ceasefire, observers saw it as a symbolic statement: the first order after the National Day – a signal that it was intended to endure.

The unprecedented measure caught global markets off guard. BBC Chinese aptly described the control order as ‘choking the neck’, a phrase that captured how the West only belatedly recognised China’s monopoly in rare earths.

But the one question endures: has the West ever necessarily understood China?

On Sunday 12 October, two days after Trump announced the 100 per cent tariff countermeasure, he posted on Truth Social: ‘Don’t worry about China, it will all be fine! Highly respected President Xi just had a bad moment’.

That same evening, markets interpreted the post as a ‘TACO trade’. Global stocks rebounded on Monday. Yet only two days later, Trump declared that the United States and China were now in a trade war.

Was it ‘just a bad moment’, or was it now in a trade war?

China’s strategic thinking has long drawn inspiration from The Art of War. In its chapter ‘Military Manoeuvres’, Sun Tzu wrote: ‘Difficult to know like yin, moving like thunder’.

As the passage suggests, when plans are concealed, they should be as impenetrable as overcast clouds, and when the force moves, it should strike with the speed of thunderbolts. Confronted with such a strategy, even the United States appears to require time to interpret and respond.

What appeared to be an abrupt, moving-like-thunderbolts announcement was in fact the result of years of endurance, of ‘sleeping on brushwood and tasting gall’, a carefully developed strategy refined over decades.

The pace and scale of this new control policy could even have led the world to misread it as a negotiating ploy, much like the reciprocal tariffs announced by the Trump administration in April, which were later delayed and diluted.

Breaking Cauldrons (Burning Bridges)

The 100 per cent tariff card played by Trump was well understood by global markets as a bargaining chip ahead of the Xi-Trump summit, unlikely to be truly implemented. The real question for the world, however, was whether China would retract its rare-earth card after the Xi-Trump meeting, just as it tactically withdrew in April.

If negotiations were to result in the United States easing its tariffs and sanctions on Chinese firms, or even partially lifting semiconductor restrictions, could China and the US, as Morgan Stanley put it, once again ‘kiss and make up’?

But rare earths are different from oil; they are not geologically rare. What China monopolises is not the resource but the technology. And unlike oil, that technology is a replicable natural asset. Nations such as the United States and others are equally capable of developing it.

In fact, ever since the West became aware of China’s alarming dominance in the rare-earth industry, several countermeasures have been set in motion. By 2025 the Trump administration had even taken direct equity stakes in rare-earth firms to rebuild an alternative supply chain.

While rare earths may not be geologically scarce, constructing a complete industrial chain requires time measured in years. According to conservative estimates by EconoFact and GIS Report, even under favourable conditions it would take the United States approximately four to ten years to develop a processing capacity of meaningful scale.

Four to Ten Years. Long or Short?

Consider China’s position. Suppose it holds a 30-year-old trump card capable of gripping both the United States and the world, one that within the next four to ten years may begin to fade, much as Japan, which once held over 80% of global semiconductor capacity, was later overtaken by the United States, South Korea and Taiwan. Would it be reasonable to fold its hand?

For a civilisation that has witnessed millennia of dynastic rivalry and long embraced the lessons of The Art of War, the answer may never be a simple ‘yes’ or ‘no’. Rather, it lies in how to maximise the strategic value of the rare-earth card before it loses its potency and becomes a discarded piece.

Another well-known maxim in The Art of War advises: ‘If your forces outnumber ten-to-one, surround them; at five-to-one, attack’. Holding nearly 90 per cent of the world’s rare-earth production capacity, China already occupies a strategic high ground where it may not need to act offensively; surrounding alone could suffice.

Even if this trump card is temporarily set aside, it will surely be played again, and many times, until the final hand is won.

At the Crossroads

In July 1945, following the Allied victory in Europe, the Allies issued the Potsdam Declaration to Japan, demanding its unconditional surrender or face ‘prompt and utter destruction’.

At the time, Japan’s government under Prime Minister Kantarō Suzuki had yet to reach a final decision on whether to accept the terms. When a reporter asked Suzuki for his stance on the declaration, Suzuki used the word ‘mokusatsu’.

The term ‘mokusatsu’ does not appear in classical Japanese. It was likely a modern wasei-kango, a ‘Japanese-made Chinese word’, combining two Chinese words: ‘silence’ and ‘erasure’.

Suzuki, a member of the peace faction, had urged the Emperor to accept the Potsdam Declaration. In this context, ‘mokusatsu’ was less aggressive than its literal meaning suggests. It was closer to a face-saving and diplomatically reserved ‘no comment’.

However, in English-language media and diplomatic channels, the word was translated as ‘ignore’ and interpreted as a categorical rejection of surrender, a mistranslation that hastened the American decision to use the atomic bomb. And the rest is history.

In truth, within the Confucian framework of East Asia, restraint in the face of one’s superior or even one’s enemy has evolved to be regarded as a virtue in modern times, and avoiding severing ties prematurely is wisdom. One does not ‘turn the face’ unless certain of holding an absolute advantage.

By 2025, rare earths had become not merely the vitamins of modern technology but the lifeblood of artificial intelligence. GPUs, servers, and cloud infrastructure all depend on rare earths. From the perspective of artificial intelligence, rare earths are the very bloodstream of the global economy.

For AI companies such as NVIDIA, tariffs and export restrictions can be managed. But the stability of raw-material supply is both the most fundamental condition and the most critical prerequisite. The global wager on AI by the United States and the world, including China, is unprecedented in scale. For the AI bubble to land softly from its clouds above, at least several years of transition will still be required.

Yet if this entire industry, built upon China’s rare-earth foundation, begins to crumble at the raw-material level, the world’s financial markets could once again face a ‘prompt and utter destruction’.

October 2025 may mark the beginning of a prolonged trade war. It is also one of the rare moments when China, facing a stronger counterpart, ‘turns its face’. Behind this gesture lies a depth of patience and resolve that Western policymakers may not yet fully comprehend.

Has the United States recognised the face behind the turn? Whether Washington once again misreads East Asia will, to some degree, determine whether 10 October 2025 will be remembered by history as a crossroads of a new era.

© Nero Huang, 2025, published by RUSI with permission of the author.

The views expressed in this Commentary are the author's, and do not represent those of RUSI or any other institution.

For terms of use, see Website Terms and Conditions of Use.

Have an idea for a Commentary you'd like to write for us? Send a short pitch to commentaries@rusi.org and we'll get back to you if it fits into our research interests. View full guidelines for contributors.