Asian AI Startups Launch Mythos-Like Models Amid Export Ban

The US government's ban on Anthropic's Mythos and Fable 5 has created an opportunity for Asian startups to launch their own AI models, with Chinese cybersecurity firm 360 and Tokyo-based Sakana AI unveiling Tulongfeng and Fugu, respectively.

The launch of these models comes as the US government's export ban on Anthropic's Mythos and Fable 5 enters its third week. The ban, which was imposed due to concerns over the potential misuse of the AI technology, has created a gap in the market that Asian startups are now looking to fill.

Sakana AI's Fugu Model

Sakana AI's Fugu model is designed to be a frontier AI model that can compete with Anthropic's Fable 5 and Mythos. The model is named after the Japanese word for blowfish and is designed to orchestrate access to other models through their APIs. According to Sakana AI, Fugu is a hedge strategy that can help preserve access to frontier AI, rather than replacing it.

• Fugu is designed to work well with small datasets and is optimized for the Japanese language and culture.
• The model is targeted at Japanese businesses and government agencies looking to reduce their exposure to tightening export controls.
• Sakana AI co-founder David Ha described Fugu as more than just a land grab during a vulnerable moment for US competitors, but rather as a way to coordinate agent usage among many models.

"Orchestration Models are the next frontier, beyond bigger models," Ha wrote on X. "Collective intelligence is the practical hedge against this concentration of power."

360's Tulongfeng Model

Chinese cybersecurity firm 360 has unveiled two AI security tools, Tulongfeng and Yitianzhen. Tulongfeng is designed to automatically discover software vulnerabilities, while Yitianzhen is built to automate cyber defense and incident response. The launch of these models comes with a message from 360's founder, Zhou Hongyi, who described vulnerability-finding AI as a national strategic asset and flagged the risk of "one-way transparency."

• Tulongfeng is designed to automatically discover software vulnerabilities.
• Yitianzhen is built to automate cyber defense and incident response.
• 360's founder, Zhou Hongyi, described vulnerability-finding AI as a national strategic asset.

Implications of the Export Ban

The US government's export ban on Anthropic's Mythos and Fable 5 has significant implications for the AI industry. The ban has created a gap in the market that Asian startups are now looking to fill, and it has also raised concerns over the potential misuse of AI technology. As the ban drags on, it is likely that more Asian startups will launch their own AI models, which could potentially challenge the dominance of US AI companies.

• The export ban has created a gap in the market that Asian startups are now looking to fill.
• The ban has raised concerns over the potential misuse of AI technology.
• The launch of Asian AI models could potentially challenge the dominance of US AI companies.