China Launches Tulong Feng in Response to U.S. AI Restrictions
In a strategic move amidst U.S. export restrictions on AI models, China has introduced Tulong Feng, a cybersecurity AI tool designed to detect vulnerabilities. The announcement was made by Zhou Hongyi, founder of Qihoo 360, at the ISC.AI 2026 conference in Beijing on June 24.
Zhou emphasized the importance of China having its own advanced AI, akin to Anthropic's Mythos, stating, "China's cybersecurity industry must have its own Mythos." Tulong Feng aims to fill the gap left by the U.S. model, which has been limited by government controls and access restrictions. Alongside Tulong Feng, Zhou also unveiled Yitian Zhen, an automated defense system, and a new security coalition called Panshi Zhidun, or Shield of Bedrock.
Highlighting the capabilities of Tulong Feng, Zhou reported that the tool has identified 3,432 vulnerabilities, with 105 validated by Chinese regulatory authorities. He compared the significance of Mythos to "cyber nuclear weapons" in AI, capable of autonomously analyzing and strategizing cyber vulnerabilities. He criticized the exclusivity of Anthropic's partnerships, emphasizing that Chinese firms lack access to collaborative initiatives like Glasswing, which includes tech giants such as Microsoft and Apple.
