The landscape of global artificial intelligence is undergoing a profound geopolitical and technological transformation. Following the Trump administration’s sweeping restrictions barring non-Americans from accessing advanced frontier AI models, international competitors have rapidly accelerated their own sovereign capabilities. Within a single week, major technological entities in both Japan and China unveiled powerful new systems explicitly positioned to rival or match the restricted American technology.
In Tokyo, AI startup Sakana AI launched "Fugu," a frontier model engineered to orchestrate complex agent tasks across multiple platforms. Concurrently, Chinese cybersecurity powerhouse 360 introduced "Tulongfeng," an advanced security-focused AI system designed to compete directly with Anthropic’s heavily guarded Mythos model. These rapid developments highlight the unintended consequences of aggressive trade and technology controls, signaling that regional markets are moving decisively to inoculate themselves against sudden disruptions in the global tech supply chain.
The Chronology of the Export Ban and Regional Reactions
The current friction stems from a strategic policy shift enacted by the United States federal government. Roughly two weeks prior to the recent Asian unveilings, the Trump administration issued an executive directive restricting global access to Anthropic’s flagship cybersecurity model, Mythos, alongside its restricted counterpart, Fugu 5. Citing national security risks associated with advanced automated cyber capabilities, the U.S. government placed these technologies under strict extraterritorial lock-and-key, denying international enterprise and governmental partners access to models that had quickly become gold standards in digital defense.
The reaction from global markets was swift. On Wednesday, Chinese cybersecurity firm 360 announced the rollout of Tulongfeng, claiming the tool matches the capabilities of Anthropic’s restricted software. Tulongfeng is engineered to automate the discovery of software vulnerabilities, while a secondary tool, Yitianzhen, was introduced to streamline automated cyber defense and incident response.
Zhou Hongyi, founder of 360, framed the introduction of vulnerability-discovery AI not merely as a commercial product launch, but as a critical national security asset. According to reports from Reuters, Zhou warned against the dangers of "one-way transparency"—a scenario where select Western actors maintain exclusive access to advanced offensive and defensive cyber capabilities while leaving other nations technologically vulnerable.
Sakana AI’s Strategic Positioning and the Fugu Release
In Tokyo, the timing of Sakana AI’s release of its blowfish-inspired model, Fugu, drew intense international scrutiny. However, company leadership maintained that the launch schedule was entirely coincidental. Sakana spokesperson statements indicated that the underlying research for Fugu had been in development for over a year, with core principles presented at the International Conference on Learning Representations (ICLR) earlier in the spring.
Despite the coincidence, Sakana AI has leaned into the market realities dictated by the U.S. export controls. The company’s promotional materials explicitly market Fugu as a mechanism for "delivering frontier capability without the risk of export controls."
Founded in 2023 by Google alumni David Ha and Llion Jones, alongside former Mercari and Stability AI executive Ren Ito, Sakana AI has carved out a distinct niche. The startup specializes in creating cost-effective generative models optimized for smaller datasets, with a particular focus on Japanese language comprehension and regional cultural nuances. In late 2025, the company secured a substantial $135 million Series B funding round at a valuation of $2.65 billion, providing the financial runway necessary to challenge Western hegemony in specialized regional AI infrastructure.
The Philosophy of Collective Intelligence and Model Orchestration
Rather than positioning Fugu as an outright replacement for American technology, Sakana AI’s leadership has framed the model as part of a broader architectural evolution: multi-model orchestration.
In public statements shared via social media and policy essays, CEO David Ha argued that relying on a single geographic provider for foundational national infrastructure is a catastrophic risk in an era of sudden trade barriers. Ha emphasized that "access to top models can disappear overnight," making collective intelligence and distributed systems an essential hedge against the centralization of technological power.
This sentiment echoes arguments made by co-founder Ren Ito at the G7 summit in Evian, where access to artificial intelligence and cross-border export controls dominated diplomatic discussions. In an op-ed published by Project Syndicate, Ito urged Western policymakers to recognize that the primary priority for allied nations should be preserving collaborative access. He asserted that artificial intelligence should function as a shared, co-developed technology rather than a hoarded commodity.
Broader Implications for the Global AI Market
The unfolding events present complex challenges for American AI labs, which have enjoyed historic financial growth. Anthropic reported that its annualized run-rate revenue crossed $47 billion in May 2026, positioning the company near a $1 trillion valuation ahead of a planned initial public offering. However, the exact financial exposure of U.S. labs to Asian enterprise markets remains opaque.
The longer export restrictions remain in place, the greater the incentive for foreign enterprises to adopt localized alternatives. Even if geopolitical tensions eventually ease and U.S. firms seek to recapture lost market share, regional competitors like Sakana AI and 360 are establishing deep operational roots. These local models offer superior alignment with regional linguistic nuances, regulatory frameworks, and cultural contexts, making them sticky alternatives for foreign governments and corporations.
As the global AI ecosystem fragments, the era of unhindered, centralized technological dominance is being challenged by a multi-polar reality. Whether through Tokyo’s focus on model orchestration and collective intelligence or Beijing’s push for sovereign cyber-defense infrastructure, the international market is actively engineering a future where dependency on a single superpower is no longer an option.
