Deal or No Deal: The U.S.-China Tech Cold War Heads to the Negotiating Table

Air Force One touched down in Beijing this week carrying something unusual: not just a president, but a delegation that reads like the Forbes 400 crossed with a corporate earnings call. 

Elon Musk. Tim Cook. Larry Fink. Kelly Ortberg of Boeing. Jane Fraser of Citi. The CEOs of Goldman Sachs, Mastercard, Qualcomm, Micron, and GE Aerospace. More than a dozen of the most powerful executives in American business, flying together to sit across the table from Xi Jinping.

That image alone tells you something the diplomatic press releases won’t.

When the heads of America’s biggest companies need to be in the room, it means the room matters. And when the agenda includes trade, artificial intelligence, Taiwan, chip controls, and the geopolitical fallout from a shooting war in the Middle East — it means this is not a routine summit. It is a reckoning.

Here is the honest assessment of where things stand, what this meeting can and cannot accomplish, and — critically — what all of it means for investors navigating one of the most consequential inflection points in modern economic history.

The Manus Moment: A Preview of How This Goes

Before we discuss where things are headed, it is worth dwelling on where they just were.

Six weeks before this summit was scheduled, China’s National Development and Reform Commission issued a terse statement ordering Meta to unwind its $2 billion acquisition of Manus. 

Manus is the agentic AI startup that many in the technology world — including this editor — have used and followed closely. The language coming out of Beijing was striking: Chinese officials reportedly characterized the deal as a “conspiratorial” attempt to hollow out the country’s technology base.

Conspiratorial. Not a legal objection. Not a regulatory formality. A word that signals how seriously Beijing views the loss of AI talent and intellectual property to the West.

The Manus situation was messy on both sides. 

The company had already relocated its operations to Singapore, reincorporating there specifically to sidestep both American restrictions on Chinese AI investments and Chinese rules limiting domestic firms from moving IP overseas. 

Meta had already integrated Manus engineers into its team. Money had already changed hands. None of that mattered - Beijing barred the founders from leaving the country and ordered the deal reversed.

That is the posture China is bringing to this summit. Not a partner eager to compromise with its cutting-edge tech. A rival determined to protect every strategic asset it has built, by any means available.

Understanding that is the starting point for everything else.

Thirty Years of Strategic Patience

It is fashionable in American political discourse to act surprised by China’s manufacturing and technology dominance. It shouldn’t be. This did not happen by accident, and it did not happen quickly.

Over the past three decades, China executed one of the most disciplined long-term economic strategies in modern history. 

WTO accession in 2001 was the opening. What followed was a systematic climb up the value chain — from cheap labor assembly to component manufacturing to vertically integrated production ecosystems that now span everything from raw materials to finished goods. 

Made in China 2025 was not a slogan. It was a blueprint, and Beijing has largely delivered on it.

The truth is that China played by its own rules throughout. Intellectual property appropriation was not a side effect of the strategy — it was a component of it. 

The infiltration of Western research institutions, the aggressive recruitment of foreign-trained engineers, the “joint venture” requirements that amounted to forced technology transfer — all of it contributed to an industrial base that is now, across several critical sectors, the global leader.

Today, China is the world’s largest automobile market and, increasingly, its largest automobile exporter. It dominates consumer electronics manufacturing. It controls the lion’s share of rare earth processing. 

And it has done all of this while straddling a political system that is nominally communist but functionally capitalist in its industrial ambitions — with the added advantage that the state can direct capital, mandate industrial policy, and absorb losses in ways that no private market competitor can match.

The tariff strategy pursued by the Trump administration has created real pressure, and it has produced genuine commitments to reshore manufacturing in specific industries. 

But the structural dependency built over thirty years does not reverse in two. The economics of Chinese production remain formidable in the near and medium term, and the CEOs sitting in that delegation understand this better than anyone.

The AI Arms Race and the Chip Moat That Wasn’t

Then came artificial intelligence — and with it, a set of disruptions that neither side fully anticipated.

The United States moved to restrict advanced chip exports to China, betting that cutting off access to state-of-the-art semiconductors would slow Chinese AI development. It was a reasonable theory. It has not played out as hoped.

The global internet is a remarkably porous system. VPNs, third-country routing, gray market hardware — the tools available to circumvent export controls when it comes to compute are numerous and well-understood.

More importantly, Chinese AI labs have demonstrated a savvy ability to do more with less. DeepSeek’s emergence was a genuine shock to the American technology establishment — not because it was better than the leading American models, but because of how capable it was at a fraction of the compute cost. 

MiniMax, which powers the default model in my own bot infrastructure, is fast, capable, and inexpensive. Manus, before the Meta drama, was the first widely recognized autonomous agentic builder.

👉If you want deeper analysis on how the global AI race is reshaping valuations, infrastructure spending, and competitive positioning across companies like Nvidia, Microsoft, Alphabet, Tesla, and Amazon, explore the Evervests AI & Technology archive and Premium Research Reports.

The Taiwan Dependency

The gap between American and Chinese AI capability is real. But with each new model iteration, it narrows. And the Chinese are not standing still.

Meanwhile, the chip restriction strategy has had unintended consequences: in addition to forcing the CCP to create their own domestic chip supply, it has made Taiwan the single most geopolitically consequential territory on earth. 

Taiwan Semiconductor Manufacturing Company produces the advanced chips that power essentially every frontier AI system. 

Nvidia’s valuations, Tesla’s autonomous vehicle ambitions, and the entire buildout of AI infrastructure depend on TSMC’s continued operation. Any scenario that disrupts Taiwan — military, economic, or otherwise — sends shockwaves through the global technology economy that dwarf anything the tariff conflict has produced.

👉 This is one of the core themes we monitor at Evervests: how geopolitical risk, semiconductor concentration, and AI infrastructure dependency could reshape long-term valuations across the technology sector over the next decade.

This is the central risk embedded in every technology-adjacent investment thesis.

Iran, Hormuz, and the Leverage Nobody Expected

The summit was originally planned for late March. It was delayed by a war.

The United States and Israel launched large-scale strikes on Iran in late February. The conflict spread across the Middle East. A fragile ceasefire has been announced, but the Strait of Hormuz — through which roughly 20% of the world’s oil flowed in peacetime — remains a pressure point especially for China. 

China is the world’s largest purchaser of Iranian oil, with its purchases accounting for an estimated 90% of Iran’s exported crude. 

That dependency gives Beijing significant influence over Tehran’s behavior — influence that Washington badly wants China to exercise right now.

Trump has reportedly been pushing Xi to use that leverage to help reopen the Strait. Xi, for his part, has leverage of his own: every day the Strait remains disrupted is a day China’s economy absorbs higher energy costs.

The incentives are not perfectly aligned, but they are not opposed either. This is the kind of overlapping interest that can produce real, if limited, cooperation.

What it also does is complicate the chip and technology conversations. When you need someone’s help on a Middle East crisis, you negotiate differently on semiconductor export controls.

What Actually Happens at Summits Like This

Leader-to-leader summits almost always improve the tone of a relationship. The atmospherics get better. Statements get issued. Both sides claim wins. 

There are genuine areas where near-term deals are possible: agricultural exports, jet engines, perhaps some AI cooperation carve-outs on non-sensitive consumer applications. Xi and Trump have demonstrated that they can reach transactional agreements when the political incentives align.

But structural change is a different matter. 

China’s Five-Year Plans — including the latest, published this past March — make unambiguously clear that Beijing’s overriding priority is self-reliance. The Made in China 2025 strategy, the China Standards 2035 initiative, the Dual Circulation framework — these are not negotiating positions. They are the architecture of CCP industrial policy. 

They do not move in a weekend of bilateral talks.

History is instructive. In 2017, China made commitments to open its market to American credit cards. Years later, execution remained slow and partial. The pattern — agreement, delay, dilution — has repeated often enough that experienced China hands price it into every assessment of summit outcomes.

Trump is a dealmaker, and he will push hard for visible wins. Xi will likely offer enough to make the trip look successful. 

Both leaders will emerge to applause. Some of those deals will matter. Most of the hard structural issues — Taiwan arms sales, chip export controls, genuine technology decoupling — will be deferred to the next meeting, or the one after that.

The Robotics Wildcard and the Supply Chain Problem

The medium-term picture is where things get genuinely interesting — and genuinely uncertain.

China faces real structural headwinds. Its population is declining, the demographic dividend from its manufacturing workforce has peaked, and energy dependency on foreign petrol remains a strategic vulnerability.

These are not trivial problems. A nation that has built its economic model on abundant, disciplined human labor faces a long-term productivity ceiling.

The push into robotics is the obvious response, and China has moved aggressively there. 

But here is the thing: robotics works both ways. Universal robotics capability, if it develops broadly, could substantially boost domestic production in the United States and elsewhere. 

If you can build a factory that runs largely on machines, the wage differential that made Chinese manufacturing so competitive starts to matter less. That is the bull case for American reindustrialization over the next decade.

👉Robotics, automation, and AI-driven manufacturing are one of the most important long-term investment themes. Evervests Premium members receive ongoing research covering the companies positioned to benefit from an autonomous future.

The problem is the vertical supply chain underneath those factories. 

Raw materials. Specialized components. Processing capacity. The rare earth supply chain is effectively a Chinese monopoly at the moment.

Advanced battery production. The intermediate goods that feed advanced manufacturing. The United States cannot currently build a self-sufficient industrial ecosystem for critical technology sectors — not in two years, not in five. Whether it can do so in ten is an open question, and the honest answer is: probably partially, for select industries, under the best-case policy scenario.

That is the real strategic gap. Not the AI models themselves, where American innovation remains formidable. Not the capital markets, where the United States retains a structural advantage. The gap is in the physical manufacturing stack — the ability to produce, at scale and at cost, the hardware that the AI and technology economy runs on.

What This Means for Investors

The investment implications are significant — and largely underappreciated by retail investors.

Semiconductor and AI infrastructure plays carry embedded geopolitical risk. 

Nvidia’s valuation, to take the most obvious example, is substantially premised on continued access to TSMC’s fabrication capacity and continued dominance in AI training hardware. Both of those assumptions carry Taiwan-related tail risk that is rarely priced into near-term earnings models. That does not mean you sell Nvidia. It means you understand what you own.

Supply chain exposure is a hidden variable in most manufacturing and consumer electronics names. 

Apple’s Tim Cook is on this trip for a reason. Apple’s manufacturing base remains deeply intertwined with Chinese production partners. Progress toward diversification into India and Vietnam is real but partial. Any significant deterioration in U.S.-China trade relations creates margin pressure at companies that have not completed — and may never complete — their supply chain diversification.

The rare earths and critical minerals story is under appreciated. 

China’s dominance in rare earth processing is one of the least-discussed strategic vulnerabilities in American technology policy. Companies and ETFs with exposure to non-Chinese rare earth development and processing — in the United States, Australia, and allied nations — represent a long-term structural theme that the current political environment is accelerating.

Robotics and automation are long-term secular themes regardless of the geopolitical outcome. 

Whether U.S.-China relations improve or deteriorate, both sides are investing heavily in automation to address labor constraints and manufacturing economics. Companies building the infrastructure for domestic robotic manufacturing — and the AI systems that run them — are positioned well across multiple scenarios.

The defense and aerospace sector has a direct line to every conversation happening in Beijing right now. 

Arms sales, Taiwan, the Iran conflict, the Strait of Hormuz — Boeing’s Kelly Ortberg is in that delegation for reasons that go beyond commercial aviation. Defense-adjacent technology companies with government contract exposure have a more complex but potentially more durable earnings profile than pure commercial tech plays in this environment.

Be skeptical of easy narratives in both directions. 

The “decoupling is complete” narrative overstates how quickly two deeply intertwined economies can separate. The “this summit fixes everything” narrative ignores thirty years of structural dynamics that two leaders cannot rewrite in a weekend. The most honest portfolio positioning acknowledges that this relationship will remain contested, interdependent, and occasionally volatile — probably for the rest of this decade and beyond.

👉 The Evervests Portfolio is designed around exactly these types of long-duration structural themes — artificial intelligence, automation, infrastructure, energy, defense, and the evolving global supply chain. Members can follow current positioning, research updates, and long-term investment theses in real time.

The Closing Assessment

Here is where we land.

The United States still holds a good hand. The foundational AI research advantage is real. The capital markets infrastructure is unmatched. The dollar system, whatever its long-term challenges, remains the backbone of global trade. 

The innovation culture — the ability to take a breakthrough and commercialize it at scale — is still, on balance, an American comparative advantage.

But the table is crowded now in a way it wasn’t twenty years ago. 

China has spent three decades building aces, and several of them are sitting in Asia. The technology race has introduced a genuine wildcard that neither government fully controls. The physical manufacturing stack remains deeply tilted toward Chinese production capacity. 

And the Taiwan risk is not theoretical — it is the central variable in any honest long-term technology investment thesis.

Trump and Xi will shake hands. Both will declare progress. Some of it will be real, in narrow, transactional ways. The structural forces will continue regardless of what is said in the Great Hall of the People this week.

The investors who navigate this well will be the ones who understand the difference between the theater of summits and the slow-moving tectonic forces underneath them. Between what gets announced and what actually changes. Between the short-term tone and the decade-long arc.

We used to hold all the cards. Now we still have a strong hand. But this is a different game now — and anyone telling you otherwise is not paying attention.

Thank you for reading.

👉For serious investors looking to understand how dramatic technological changes will affect investing, become a premium member at Evervests.com.

Images: Unless specifically noted, all images were created by Evervests.com using AI.

Disclaimer: The content above represents the opinion of Evervests and is provided for informational purposes only. It does not constitute investment advice, financial advice, or any other professional advice. Readers should conduct their own research and consult with qualified professionals before making any financial decisions. Past performance is not indicative of future results. Evervests makes no representation as to the accuracy, completeness, or reliability of any information provided herein.