Asian AI Models Rise as US Export Ban Creates Gap

Asian AI startups are launching new models to fill a gap created by a US export ban. This ban prevents non-Americans from accessing Anthropic's Mythos and Fable 5 models. Here's the thing: Two new Asian models are emerging that promise similar capabilities. Chinese cybersecurity firm 360 unveiled Tulongfeng, which they say can compete with Anthropic's cybersecurity-focused Mythos. China 360's founder, Zhou Hongyi, states that powerful AI tools should not be exclusively held by others. Earlier in the week, Tokyo-based Sakana AI launched Fugu. This model, built for agents, is designed to stand alongside Fable 5 and Mythos Preview. Sakana AI says Fugu can orchestrate access to other models through their APIs. What's interesting is that both launches come as the US ban enters its third week. Sakana AI claims the timing is coincidental, and their research on Fugu predates the ban. However, their website highlights frontier capability without export control risks. The bottom line is this creates significant stakes for Anthropic, whose run-rate revenue reached $47 billion in May 2026. Local rivals, trained for regional language and nuance, are already moving into this space.