Asian AI companies are rushing to fill the gap left by the US export ban on Anthropic’s most powerful models. Tokyo-based Sakana AI released Fugu, an orchestration model it claims matches Fable 5 on key benchmarks, while Beijing cybersecurity firm 360 Security unveiled Tulongfeng, a vulnerability-discovery tool it says rivals Mythos. Both launched as the ban entered its third week with no resolution in sight.
The rapid response shows how quickly competitors move when a market leader is forced to step back. The story extends our earlier coverage of the Mythos and Fable 5 export ban and its global fallout.
Sakana Fugu and 360 Tulongfeng Step In
As TechCrunch reported, Sakana AI’s Fugu targets the orchestration niche that Fable 5 occupied, while 360 Security’s Tulongfeng aims squarely at the cybersecurity capabilities that made Mythos so valuable and so tightly controlled. Both companies are positioning their tools as ready substitutes for customers cut off from Anthropic.
The choice of a cybersecurity-focused rival is notable. Mythos was restricted precisely because of its powerful security capabilities, and a Beijing firm openly building an equivalent underscores how export controls can accelerate the very competition they aim to contain.
How the Mythos Export Ban Opened the Door
The Trump administration banned Mythos and its more restricted version, Fable 5, from non-American hands, citing national security concerns over the models’ cybersecurity power. With Anthropic forced to disable foreign access, customers outside the US suddenly needed alternatives, and regional labs moved fast to supply them.
This dynamic illustrates a recurring risk of export controls: they can hand domestic markets to foreign challengers. The US recently allowed limited Mythos access for trusted partners, as we covered in the trusted-partner release, but the foreign window had already opened.
A Wave of Regional Challengers Emerges
Sakana and 360 are not alone. As TheNextWeb reported, Singapore-based Vertex AI launched its Phoenix-7 model promising Mythos-level reasoning without the Washington red tape, while South Korean unicorn Mindforge unveiled Atlas, claiming parity with Mythos on enterprise benchmarks. Within weeks, a cluster of credible alternatives emerged across Asia.
The breadth of the response suggests the export ban did not slow global AI capability so much as redistribute it. Multiple regions now host models marketed as Mythos equivalents, fragmenting what had been a US-led lead in frontier AI.
Why Local Models May Keep the Lead
Even if US firms eventually regain access, the regional alternatives carry a durable advantage: they are trained to better understand local languages and cultural nuance. That localization could keep customers loyal long after any ban ends, as we noted in our coverage of the broader AI model race.
For Anthropic, the lesson is sobering. A temporary restriction may produce permanent competition, as customers who switch to capable local models have little reason to switch back once they are embedded in regional workflows and toolchains.
Related Articles
Anthropic Fable 5 and Mythos 5 Export Ban Explained
Trust Post Desk
A journalist and editor at TrustPost.org covering world and national news, technology updates and human-interest stories. They check every fact, interview sources in person or online, and aim to deliver clear, accurate reporting. Their work ranges from breaking news to in-depth features and daily newsletters. Outside the newsroom, they follow emerging trends and engage with readers on social media.
