Asian AI startups launch Mythos-like models as Anthropic export ban continues

On Wednesday, Chinese cybersecurity company 360 reportedly launched Tulongfeng, an AI tool that it says can directly confront Anthropic’s Mythos. That is a cybersecurity-focused AI model that is reportedly so powerful, the Trump Administration has currently banned it and its more restricted version, Fable 5, from the hands of non-Americans.

Earlier in the week Sakana AI, a Tokyo-based AI startup launched Fugu, a model named after the Japanese word for blowfish. The company says this frontier AI model “stands shoulder to shoulder with leading models like Anthropic’s Fable 5 and Mythos Preview.” It is also designed for agents, who have the ability to organize access to other models though their APIs.

Two new products for Asian models come as the US government ban continues. The order preventing Anthropic from worldwide access to Mythos and Fable took place two weeks ago.

A spokesperson for Sakana AI told TechCrunch that the release of its new model was “totally coincidental,” but that hasn’t stopped it from spending just yet. Its website advertises “delivering border energy without the risk of export controls.”

“Sakana Fugu is something we’ve been building since last year – the research behind it was presented at ICLR this spring, and it shows a key step in how we’re bringing frontier-level value to Sakana AI. We were confident in the product’s merits; the timing just happened to bring more attention than we expected,” a spokesperson said of the launch of the My/F during the release.

Sakana, founded in 2023 by former Google researchers Ren Ito, Lion Jones and David Ha, makes affordable AI generative models that work well with small data sets and are optimized for the Japanese language and culture.

While the company is targeting Fugu at Japanese businesses and government agencies looking to reduce exposure to tightening export controls, it has not announced a permanent move to US AI in Asia.

“American models are still important in Asia,” the spokesman said, a view echoed by founder Ren Ito at the G7 summit in Evian last week, where access to AI and export controls were among the main topics. “We’ll show the current season in those terms rather than making permanent changes to any set of players.”

Sakana founder Ren Ito elaborated on that idea in an op-ed published in Project Syndicate last week. He urged the US federal government to consider that “its priority should be to maintain access,” to America’s closest allies, and said that “AI should not be a collective technology; it should be one that is developed together.”

David Ha, founder and CEO of Sakana, described Fugu as more than just a land grab at a time when US rivals are at risk. It is designed to coordinate the use of agents between multiple models.

“Orchestration models are the next frontier, beyond big models,” he wrote in X. Relying on a single national infrastructure provider, he argued, is a risk that recent export controls have made it impossible to ignore.

“Access to top models can disappear overnight,” he wrote. “Collective intelligence is an effective defense against this concentration of power.”

While Tokyo-based Sakana positioned Fugu as a hedge strategy, a last resort to reach the AI ​​frontier, not a replacement, China’s 360 was unstoppable.

A Chinese company has reportedly launched two AI security tools. Tulongfeng is designed to automatically detect software vulnerabilities, and Yitianzhen is designed to automate cyber security and incident response.

The launch of the product, however, came with a message. According to Reuters, 360 founder Zhou Hongyi described risk detection AI as a national strategic asset, and flagged what he called the risk of “one-way visibility”, a situation where some players cannot access advanced risk detection capabilities while others cannot.

Anthropic had been on an epic growth trajectory. The US AI lab said its run-rate revenue will exceed $47 billion by May 2026. How much of that depends on Asian business customers is not publicly known.

But in the weeks since the export order went into effect, at least two companies, one in Tokyo, one in Beijing, have moved into the space it left behind. Even if US companies cannot regain trust if this ban ends, local alternatives, trained to better understand the local language and nuances, are already closing the gap.

360 did not respond to a request for comment.