Asian AI startups launch Mythos-like models as Anthropic’s export ban drags on
TL;DR
China’s cybersecurity company 360 unveiled Tulongfeng, an AI tool for automated vulnerability discovery that it claims can compete with Anthropic’s Mythos. Tokyo-based Sakana AI launched Fugu, an orchestration model for agents that can coordinate access to other models through APIs and is aimed at Japanese companies and government buyers. The launches follow a U.S. order that blocks Anthropic from giving non-Americans global access to Mythos and the more restricted Fable 5.
Nauti's Take
The U. S.
export-control play is meant to preserve leverage, but it also creates the market it wants to slow down. Every restriction becomes a sales pitch for local providers: similar direction, less dependency, better political fit.
Sakana’s messaging still carries some opportunistic PR timing, while 360’s positioning is closer to national cyber strategy. The practical lesson for buyers is blunt: an AI stack tied to one country, one vendor, and one permission layer is not a resilient stack.
Briefingshow
This is bigger than one Asian alternative model. If companies and governments fear that access to a U. S.
model can disappear overnight, availability becomes a core buying criterion. Local providers can then compete on political reliability, not just price or language fit. For U.
S. labs, lost trust in enterprise and government deals may be harder to win back than a feature gap.
Sources