While Washington spent the last year pulling down its digital window shades, Beijing just threw open the front doors to the future.
It is a quiet, calculated masterstroke. While the United States retreats into a protectionist crouch, viewing artificial intelligence purely as a national security asset to be guarded and fenced, China is presenting itself as the ultimate ally to the rest of the world.
If you want to know how global power shifts, you don't just look at who has the fastest microchips. You look at who shows up to write the rules. Right now, that's China.
The stark reality became obvious at the United Nations first Global Dialogue on AI Governance in Geneva. Over 4,000 delegates from 170 countries packed into the Palexpo convention center. They were looking for answers. The threat of automated chaos, digital divides, and corporate monopolies hung heavy in the air. The West offered warnings, strict categories of risk, and export controls.
China, instead, offered a checkbook, an open invitation, and free code.
The Vacuum Washington Left Behind
Let's look at how we got here. In January 2025, Donald Trump hit the ground running by revoking the previous administration's AI executive order. That old order was far from perfect, but it tried to build some unified domestic guardrails. The new directive from Washington was simple: kill the red tape, strip away state-level regulations, and focus entirely on absolute American dominance.
American policy became deeply inward-looking. The goal is to let Silicon Valley builders run fast, outpace foreign rivals, and lock down proprietary technology. To the rest of the world, Washington's message felt clear: We are building the smartest systems in human history, and you cannot have them.
This left a massive space on the international stage.
The Western world treats AI regulation as an elite club. The European Union has its strict AI Act, which slaps heavy burdens on general-purpose models. The US relies on executive muscle and litigation task forces to clear domestic hurdles. Both approaches treat the Global South as an afterthought.
Enter Beijing. China looked at this fractured, values-based approach and saw an open goal. They realize that if you leave 80% of the world out of your club, those countries will look for another one.
The Rise of World AI Cooperation Organization
China didn't just show up to Geneva to talk. They brought an entirely separate, parallel institutional framework that they have been building for a year.
Back in July 2025, Chinese Premier Li Qiang laid out a 13-point international plan at a conference in Shanghai. He proposed something called the World AI Cooperation Organization, or WAICO. Fast forward to June 2026, and Beijing dropped a comprehensive white paper confirming they are fast-tracking this new organization. The headquarters are already going up in Shanghai.
Think about the sheer political weight of that. WAICO isn't designed to be a Western-led alliance like the G7 or an exclusive regulatory block like the EU. It is explicitly marketed as an open-door club. The core philosophy treats AI as a development issue, not a weapon.
To a developing nation in Latin America, Africa, or Southeast Asia, the two pitches look entirely different.
- The American Pitch: "Our companies have the best models. You can buy access via API, but we might cut you off if our security goals change. Also, don't buy Chinese hardware."
- The Chinese Pitch: "Join WAICO. We will respect your digital sovereignty. We won't force you to pick a side. Let's build local data centers together and use open models to upgrade your schools and hospitals."
Honestly, it is not a hard choice for a country struggling with basic internet connectivity.
Weaponized Open Source and Shifting Software Economics
The strategy goes much deeper than diplomatic papers. It lives in the code itself.
For the past couple of years, Western labs like OpenAI and Anthropic built deep moats around their frontier systems. They sell access. They hide weights. They charge per token. It is a brilliant business model, but it requires everyone else to remain a customer forever.
Meanwhile, Chinese labs have flooded the market with highly capable open-weight models like Qwen and DeepSeek. These models are remarkably cheap to run. In some benchmarks, they run circles around Western models that cost ten times as much to train.
By pushing open-source models, China changes the entire economic calculus of the industry. It systematically undercuts the return on investment for US software companies. If a developer in Jakarta or Nairobi can download a top-tier Chinese model for free and run it on their own servers, why would they pay subscription fees to an American corporate giant?
During the UN sessions in Geneva, Chinese delegates explicitly used this to score points. They stood up and declared that open-source AI is a shared asset for all humanity. It sounds incredibly noble. It sounds democratic.
But it is also brilliant geopolitics. It makes the world dependent on the Chinese open-source ecosystem while simultaneously starving Western venture-backed giants of international revenue. It is a brilliant play.
The Myth of Neutral Technology
Of course, we shouldn't fall for the marketing completely. China’s push for "digital sovereignty" is deeply tied to its domestic political goals.
When Beijing talks about a country's right to independently choose its AI products without being coerced, they are also saying that every government has the right to control information within its borders. Their models are heavily scrubbed to comply with strict state ideology. An AI system that mirrors authoritarian norms is a feature, not a bug, for many regimes around the world.
The Western model of the internet was built on democratic values and individual liberties. It allowed Western tech firms to dominate the globe for three decades. China wants to replace that old system with an approach where state control is the baseline rule.
They are backing up this vision with real resources. At the UN dialogue, Li Lecheng, China's minister of industry and information technology, pledged that Beijing will run 200 specific digital economy and AI training programs for Global South nations over the next five years. They are building the China-BRICS AI Development and Cooperation Center in Shanghai. They are launching the China-Laos AI Innovation Cooperation Center.
They are creating dependency through education, training, and infrastructure.
Concrete Steps for Western Leaders
The current strategy of relying solely on raw compute power and export bans is failing. If the West wants to maintain influence over how the world uses artificial intelligence, it needs to change its play.
Shift from security to development
Stop treating international AI dialogues as counter-espionage briefings. Western foreign policy needs to offer real, tangible infrastructure support to developing nations. If you don't help them build local data centers, they will let China build them.
Support Western open architecture
Clinging exclusively to proprietary, closed-door models leaves the entire open-source world to be shaped by Chinese engineers. Western governments should incentivize and fund open-weight projects that embed open, transparent values into the foundation of global software.
Show up to the multi-lateral table
You can't write the rules if you aren't in the room. Retreating from international bodies like the UN or dismissing global dialogues as talk shops just hands the microphone to Beijing. It allows them to position themselves as the sole champions of the developing world.
The race for AI supremacy isn't just happening in server farms in Iowa or labs in San Francisco. It is happening in committee rooms in Geneva, in training centers in Vientiane, and in the open-source forums where the global developer community actually works. Right now, Washington is winning the race to build the biggest models, but Beijing is winning the race to sign up the rest of the world.
