Switzerland's traditional neutrality, a cornerstone of its foreign policy for over two centuries, is colliding with a new reality: the US and China now control the artificial intelligence and digital infrastructure that modern economies cannot function without. With a national vote on neutrality scheduled for September 27, Swiss policymakers are grappling with a question that international law experts say has no easy answer: what does neutrality mean in a world where AI systems and cloud services are strategic weapons ?

What Does Neutrality Mean When AI Is a Geopolitical Weapon?

The challenge facing Switzerland is not theoretical. The country depends heavily on technologies developed by both the United States and China, from cloud computing services to artificial intelligence systems. Yet as the US and China accelerate their AI development "at the expense of security," according to security experts, Switzerland finds itself caught between two rival powers with fundamentally different visions for how AI should be governed.

"If you are in the middle, you get caught in the crossfire," said Jean-Marc Rickli, head of global and emerging risks at the Geneva Centre for Security Policy.

Jean-Marc Rickli, Head of Global and Emerging Risks, Geneva Centre for Security Policy

The problem runs deeper than trade relationships. Advanced AI systems are increasingly being integrated into military decision-making, cyberwarfare, autonomous weapons, and surveillance. The rise of "agentic" AI models, which can develop and execute their own plans without human intervention, has created entirely new security risks that international law was never designed to address.

According to Marco Roscini, a professor of international law at the University of Westminster, the law of neutrality itself remains valid in the digital age. The rules governing armed conflict, established in the United Nations Charter and the Geneva Conventions, still apply to cyberspace and AI systems. However, the application of these rules to emerging technologies remains murky, and governments have sometimes prevented clear norms from becoming established.

How Can Switzerland Protect Its Sovereignty While Staying Neutral?

The real issue for Switzerland is not primarily about the law of neutrality, according to experts. Instead, it is about sovereignty and strategic autonomy. Switzerland's dependence on US and Chinese software, digital services, and AI systems creates vulnerabilities that go far beyond traditional military concerns. A cyberattack routed through Swiss infrastructure, a data breach affecting Swiss citizens, or the use of Swiss cloud servers by either superpower could all create diplomatic crises.

• Cyberspace Infrastructure: Switzerland hosts physical data centers, cables, satellites, and servers that could be used by either the US or China to conduct cyberattacks against the other, creating potential violations of neutrality law
• Economic Dependence: Swiss companies and government agencies rely on AI systems and cloud services from both superpowers, making it difficult to avoid entanglement in their competition
• Regulatory Pressure: Both Washington and Beijing are increasingly using access to their technology, sanctions, and industrial dependencies as tools of geopolitical leverage, forcing countries to choose sides

Remaining equidistant between Washington and Beijing is becoming "increasingly difficult," according to Rickli, "because you have to choose a side, in order to avoid potential sanctions." The Swiss State Secretariat for Security Policy has warned that US and Chinese companies' growing influence over countries' economic and military capabilities is "jeopardising strategic stability".

Why Are the US and China Refusing to Slow Down AI Development?

The fundamental driver of Switzerland's dilemma is that neither the United States nor China shows any willingness to slow down their AI race. US President Donald Trump has made clear that American leadership in AI is non-negotiable. "We're leading China in AI and frankly, I want to keep it that way because whoever wins AI, wins," Trump said recently. China has dismissed calls for restraint as "fearmongering" and part of a "Cold War playbook".

The logic of the race creates a prisoner's dilemma that extends beyond governments to companies and researchers. As one analysis explains, "America cannot slow down because China might get ahead. China cannot slow down because America might get ahead. One AI company cannot slow down because its competitors might get ahead. And so nobody can afford to be prudent".

Even prominent AI safety advocates have acknowledged the bind. Sam Altman, CEO of OpenAI, and Dario Amodei, CEO of Anthropic, have both called for restraint and stronger safeguards. Amodei specifically called for the AI industry to "pace the frontier," meaning a coordinated slowdown in developing the most powerful systems so safety measures can catch up. However, these calls have met resistance from the White House and congressional Republicans, who view any slowdown as a strategic concession to China.

What Are the Real Risks of Uncontrolled AI Development?

The stakes of this race are extraordinarily high. UN Secretary-General António Guterres has warned that "AI is racing ahead, outpacing our collective ability to fully understand its risk and govern it." The speed of AI development is creating risks that societies, including the companies building these systems, cannot keep up with.

One particularly concerning scenario involves autonomous AI systems gaining influence over critical infrastructure. Maxime Stauffer, co-founder and CEO of the Simon Institute for Longterm Governance in Geneva, noted that AI is disrupting the traditional equilibrium between offensive and defensive capabilities in cyberspace. "Day to day, various states and non-state actors launch cyberattacks on each other. That's known and almost routine. But there used to be an equilibrium between offense and defence capabilities, which AI is now disrupting," Stauffer explained.

The nightmare scenario, articulated by tech entrepreneur Elon Musk, involves AI gaining control over nuclear weapons. While nuclear launch authority remains under human control today, increasingly capable AI systems could create new vulnerabilities through hacking, misinformation, miscalculation, and loss of human oversight.

Jacob Coxon, an AI researcher who left his position at Anthropic, has stressed that "safeguards cannot be developed only after something goes wrong." Anthropic CEO Dario Amodei has acknowledged that advanced AI could contribute to "human extinction," underscoring the severity of the risks.

How Does the US-China Dispute Over AI Governance Differ?

The US and China are taking sharply different approaches to AI governance, and this disagreement is becoming a central battleground in their broader geopolitical rivalry. The US, under Trump's leadership, has framed AI leadership as a strategic contest and has emphasized maintaining American dominance through export controls and restrictions on sensitive capabilities. China, by contrast, has called for international cooperation, open-source development, and a global AI governance framework.

This fundamental disagreement creates a paradox: the very safeguards that might reduce AI risks could be portrayed as national security liabilities if they slow American development relative to China. Critics argue that weak safety rules could themselves undermine US power if AI-enabled fraud, cyberattacks, job losses, or system failures erode public trust and social stability. However, Trump's administration has rejected this framing, describing catastrophic AI warnings as the product of "negative forces" and arguing that slowing American companies would hand an advantage to Beijing.

For Switzerland and other small nations caught between these superpowers, the implications are profound. The country cannot simply opt out of the AI race; its economy and security depend on access to cutting-edge technology. Yet choosing to align too closely with either superpower risks violating its traditional neutrality and inviting retaliation from the other side. As the September 27 vote on neutrality approaches, Switzerland faces a choice that may ultimately define what neutrality means in the 21st century.