The US government's export ban on Anthropic's most advanced AI models has sparked a flurry of activity among Asian AI startups, with two companies launching products that position themselves as alternatives to the banned Mythos and Fable 5 models.

What Happened

This week, Tokyo-based Sakana AI released Fugu, an orchestration model that matches Fable 5 on key benchmarks, while Beijing cybersecurity firm 360 Security unveiled Tulongfeng, a vulnerability-discovery tool that claims to rival Mythos. The launches come as the US government's export ban on Anthropic's most capable models entered its third week with no resolution in sight.

Sakana AI's Fugu is an unusual approach to building frontier-level AI models. Rather than training a new model from scratch, Sakana built a seven-billion-parameter orchestrator that decides which external model should handle each part of a problem. The company claims that Fugu routes tasks across a pool of available models, assembling and coordinating them as a team, with results matching the performance of systems that cost orders of magnitude more to train.

Background and Context

The export ban on Anthropic's most advanced AI models has been in place for two weeks, following an order from the US government. The directive prevents foreign nationals from accessing Fable 5 and Mythos, citing national security concerns. Anthropic has stated that it will "abruptly disable" its most advanced AI models for all users as a result.

The ban has sparked debate over the status of artificial intelligence (AI) gap between the US and China. Chinese AI companies have been quick to capitalize on the situation, with Sakana AI's co-founder Ren Ito arguing that relying on a single provider for national infrastructure is a risk made impossible to ignore by the export ban.

China's 360 Security has also taken advantage of the situation, unveiling Tulongfeng and Yitianzhen at the ISC AI 2026 cybersecurity conference in Beijing. The company claims that Tulongfeng can identify software vulnerabilities with ease, while Yitianzhen is designed to automate cyber defence and incident response.

Why It Matters

The launch of Sakana's Fugu and 360 Security's Tulongfeng has significant implications for the adult industry. The ability to access advanced AI models without relying on a single provider can provide greater flexibility and scalability for platforms and operators.

The export ban on Anthropic's most capable models also highlights the growing tension between AI developers and regulators over how to assess risks from so-called "jailbreaks", or methods used to bypass model safeguards. This has significant implications for industries that rely heavily on AI, such as adult content platforms.

What Comes Next

The launch of Sakana's Fugu and 360 Security's Tulongfeng marks a turning point in the development of AI models outside of the US. The question remains whether these companies can replicate the performance of Anthropic's banned models, and what implications this will have for the industry as a whole.

Key Facts

  • Sakana AI released Fugu, an orchestration model that matches Fable 5 on key benchmarks.
  • Beijing cybersecurity firm 360 Security unveiled Tulongfeng, a vulnerability-discovery tool that claims to rival Mythos.
  • The US government's export ban on Anthropic's most capable models entered its third week with no resolution in sight.
  • Anthropics will "abruptly disable" its most advanced AI models for all users as a result of the export ban.
  • China's 360 Security claims that Tulongfeng can identify software vulnerabilities with ease, while Yitianzhen is designed to automate cyber defence and incident response.