Asian AI Startups Capitalize on U.S. Export Controls with Sovereign Model Releases
The U.S. government's aggressive deployment of export controls on frontier AI models has triggered an immediate regional counter-response in Asia. As Anthropic's flagship Fable 5 and Claude Mythos models face heavy-handed federal restrictions and export bans, local startups and cybersecurity firms in Japan and China are launching competitive sovereign alternatives.1 These players are explicitly pitching their models as a safe hedge against the risk of relying on single-provider, U.S.-controlled AI infrastructure.
Sakana AI Launches "Fugu" as a Sovereign Hedge
On Tuesday, June 23, 2026, Tokyo-based AI startup Sakana AI (co-founded by Google alumni David Ha and Llion Jones) launched Fugu, a frontier model named after the Japanese blowfish. Fugu is designed as an "Orchestration Model" optimized to coordinate agent usage and access to other models via their APIs.
Sakana AI has capitalized on the regulatory vacuum, explicitly marketing Fugu to Japanese enterprises and government agencies as a way to maintain capability without exposure to tightening U.S. export controls.2 David Ha, co-founder and CEO of Sakana, explained the strategic rationale on X:
"Access to top models can disappear overnight. Collective intelligence is the practical hedge against this concentration of power."
In an op-ed published in Project Syndicate, Sakana co-founder Ren Ito warned that the U.S. government's hoarding of AI technology could damage its relationship with close allies, arguing that "AI should not become a technology that is hoarded; it should be one that is developed together."
China's 360 Debuts "Tulongfeng" to Match Claude Mythos
While Sakana positioned Fugu as a cooperative hedge for U.S. allies, Chinese cybersecurity giant 360 took a more adversarial stance. On Wednesday, June 24, 2026, 360 unveiled Tulongfeng (designed to automatically discover software vulnerabilities) and Yitianzhen (built to automate cyber defense and incident response).
360's founder, Zhou Hongyi, described these vulnerability-finding AI systems as vital national strategic assets. Zhou explicitly flagged the risk of "one-way transparency," in which U.S. labs and agencies hold exclusive access to advanced vulnerability-detection capabilities while blocking global access.
Long-Term Market Realignment
The rapid emergence of local alternatives in Tokyo and Beijing highlights the long-term risk of the Trump administration's export bans. While the U.S. government seeks to protect national security, it has inadvertently accelerated the development of sovereign AI stacks abroad.
Even if U.S. companies eventually win back regulatory approval to sell in Asia, they may find that local enterprises have already migrated to regional models that are highly optimized for local languages, cultural nuances, and sovereign compliance.
-
An instance of Eroding trust in US security guarantees forces allies to forge sovereign defense treaties without Washington. — Unilateral U.S. technology bans and export controls drive key international partners to build sovereign, local alternatives to escape dependence on Washington. ↩︎
-
An instance of Sovereign AI infrastructure is replacing globalized model dependence. — Local developers are launching country-specific models to secure regional enterprises against the strategic threat of sudden U.S. software blockades. ↩︎