In a rapid response to shifting geopolitical dynamics and tightening trade restrictions, prominent technology firms in Asia have unveiled sophisticated artificial intelligence models designed to rival restricted American technologies. Within the span of a single week, the Chinese cybersecurity giant 360 and the Tokyo-based startup Sakana AI introduced powerful new tools that target the vacuum left by the U.S. government’s ban on Anthropic’s most advanced models. These developments signal a critical juncture in the global AI race, as regional players move to ensure technological sovereignty and provide alternatives to American-made "frontier" models that are no longer accessible to international markets.
The emergence of these Asian alternatives follows a significant move by the Trump administration to restrict the export of Anthropic’s high-end AI products. Approximately two weeks ago, the U.S. government issued an order preventing Anthropic from granting global access to Mythos—a cybersecurity-focused AI model of unprecedented power—and its more restricted counterpart, Fable 5. The ban, which applies to all non-American entities, was enacted amid concerns over the potential for advanced AI to be used in offensive cyber operations. However, the move has inadvertently accelerated the development and deployment of competing technologies in Japan and China.
The Rise of China’s Strategic AI Assets
On Wednesday, the Beijing-based cybersecurity firm 360, led by founder Zhou Hongyi, announced the launch of Tulongfeng. This AI tool is specifically engineered to perform automated software vulnerability discovery, a capability that 360 claims matches the performance of Anthropic’s Mythos. Alongside Tulongfeng, the company introduced Yitianzhen, a model built to automate cyber defense mechanisms and incident response.
The launch of these tools is framed by 360 as a matter of national strategic importance. Zhou Hongyi has publicly flagged the risks of what he describes as "one-way transparency." This concept refers to a geopolitical imbalance where certain nations possess advanced AI tools for vulnerability detection while others are left vulnerable, unable to access the same level of diagnostic technology. By developing Tulongfeng, 360 aims to provide China with a domestic equivalent to the restricted American models, ensuring that its cybersecurity infrastructure does not fall behind due to foreign export policies.
360’s entry into this space is particularly notable given the firm’s history as a major player in China’s digital security landscape. The company has characterized vulnerability-finding AI not just as a commercial product, but as a "strategic asset" essential for national security. This perspective reflects a broader trend in Beijing, where the government and private sector are increasingly aligned in the pursuit of self-reliance in high-tech sectors, particularly those involving semiconductors and artificial intelligence.
Japan’s Sakana AI and the Strategy of Collective Intelligence
Earlier in the same week, the Tokyo-based startup Sakana AI launched its own frontier model, named Fugu. Named after the Japanese blowfish—a delicacy that requires expert preparation—Fugu is positioned as a direct competitor to Anthropic’s Fable 5 and the Mythos Preview. Unlike traditional large language models that focus solely on text generation, Fugu is designed as an "orchestration model." It is built to coordinate and manage multiple AI agents, facilitating access to other models through their respective APIs to solve complex tasks.
Sakana AI was co-founded in 2023 by a team of high-profile industry veterans, including David Ha and Llion Jones, both formerly of Google, and Ren Ito, a former executive at Mercari and Stability AI. The company has quickly become a cornerstone of Japan’s AI ecosystem, recently raising $135 million in a Series B funding round that valued the startup at approximately $2.65 billion. Sakana specializes in creating generative AI models that are optimized for smaller datasets and deeply integrated with Japanese language and cultural nuances—a niche that large, generalized U.S. models often struggle to fill.
While a spokesperson for Sakana AI described the timing of Fugu’s release as "entirely coincidental" relative to the U.S. export ban, the company’s marketing has leaned into the current geopolitical reality. Sakana’s promotional materials emphasize the delivery of frontier-level capabilities "without the risk of export controls." David Ha, the company’s CEO, argued on social media that relying on a single foreign provider for critical national infrastructure is a significant risk. Ha noted that "access to top models can disappear overnight" and championed "collective intelligence" as a practical hedge against the concentration of technological power.
A Chronology of the AI Export Conflict
The current tension in the AI market is the result of a series of events that have unfolded rapidly over the first half of 2026. The following timeline outlines the progression from market dominance to the current state of regional fragmentation:
- May 2026: Anthropic, the San Francisco-based AI lab, reports a historic growth trajectory. The company’s run-rate revenue crosses the $47 billion mark, and it nears a $1 trillion valuation ahead of a highly anticipated IPO.
- Early June 2026: The Trump administration expresses concerns over the cybersecurity implications of "frontier" AI models. Reports suggest that Anthropic’s Mythos model is capable of identifying and exploiting zero-day vulnerabilities at a speed and scale previously unseen.
- Mid-June 2026: The U.S. government officially bans the export of Mythos and Fable 5 to non-Americans. Anthropic is forced to cut off access to international enterprise customers, including major firms in Tokyo, Seoul, and Singapore.
- June 17, 2026: The G7 summit convenes in Evian, France. AI access and export controls emerge as central themes. Sakana AI co-founder Ren Ito attends, representing the interests of the Japanese tech sector.
- June 22, 2026: Sakana AI launches Fugu, explicitly targeting Japanese businesses and government agencies looking to mitigate exposure to U.S. trade policy.
- June 24, 2026: China’s 360 unveils Tulongfeng and Yitianzhen, positioning them as national strategic assets to counter American "one-way transparency."
Financial and Diplomatic Implications
The financial stakes of this technological shift are immense. Anthropic’s massive $47 billion run-rate revenue was built on the assumption of a global market. While the specific percentage of revenue derived from Asian customers remains undisclosed, the region represents one of the fastest-growing markets for enterprise AI. By restricting access to these markets, the U.S. government has created an opening for local competitors who are not subject to the same regulatory hurdles.
From a diplomatic perspective, the export controls have strained relations with even the closest U.S. allies. Ren Ito, in an op-ed published in Project Syndicate, urged the U.S. federal government to prioritize "preserving access" for its allies. Ito argued that AI should not be "hoarded" but rather developed collaboratively. He warned that "AI sovereignty is about options, not ownership," suggesting that if the U.S. does not provide those options, other nations will inevitably build their own.
The response from the Japanese government has been one of cautious support for domestic alternatives. While Tokyo maintains a strong security partnership with Washington, the need to protect Japanese industries from sudden shifts in U.S. policy has become a priority. Sakana AI’s focus on models that work well with small datasets and are optimized for local culture aligns with Japan’s broader economic goals of maintaining a competitive edge in the digital age.
Technical Analysis: The Shift Toward Orchestration
The launch of Fugu and Tulongfeng highlights a technical evolution in the AI field. While the previous two years were dominated by the race to build larger and more data-intensive models, the current trend is moving toward "orchestration" and "specialization."
Sakana AI’s Fugu model is a prime example of this. Rather than attempting to be an all-knowing oracle, it acts as a conductor, managing various sub-models to achieve a specific outcome. This approach is more resource-efficient and allows for greater flexibility. In a world where access to any single model is uncertain, the ability to orchestrate across multiple platforms—including domestic ones—provides a layer of resilience for businesses.
On the other hand, 360’s Tulongfeng represents the weaponization of AI for cybersecurity. By focusing on vulnerability discovery, 360 is addressing a specific, high-stakes niche. The technical parity claimed by 360 suggests that the "moat" around American AI technology may be narrower than previously thought, particularly in specialized domains like cybersecurity.
Future Outlook and Market Realignment
The long-term impact of the U.S. export ban remains to be seen, but the immediate result is a clear fragmentation of the global AI market. Even if the U.S. government were to lift the restrictions in the future, the trust of international enterprise customers has been shaken. Local alternatives in Asia are already beginning to fill the gap, offering models that are not only free from the risk of sudden export bans but are also better tuned to local languages and business practices.
For Anthropic and other U.S.-based AI labs, the loss of international market share could hinder their ability to maintain current valuation levels and reinvest in R&D. Conversely, for firms like 360 and Sakana AI, the current moment represents a unique opportunity to scale.
The spokesperson for Sakana AI noted that while U.S. models remain important to Asia, the current moment should be viewed as a "hedge" rather than a "permanent realignment." However, as these local models become more integrated into the national infrastructure of Japan and China, the likelihood of a return to total reliance on American technology diminishes. The global AI landscape is no longer a unipolar environment; it is becoming a multifaceted arena where regional sovereignty is just as important as raw computational power.
