China’s 360 Security Technology unveils AI models for vulnerability discovery

China's 360 Security Technology has unveiled two AI models, collectively called Yitian Tulong, designed for vulnerability discovery and automated defense. Founder Zhou Hongyi presented the models at the ISC.AI 2026 conference, positioning Tulongfeng as a "Chinese Mythos" capable of identifying software flaws, with further coverage provided by Tech Radar.

During the ISC.AI 2026 conference in Beijing, 360 Security Technology introduced Yitian Tulong, comprising two AI models: Tulongfeng for vulnerability discovery and Yitianzhen for automated defense and incident response. Tulongfeng reportedly found 3,432 software flaws, including 105 confirmed by the Chinese government.

While acknowledging a 20-30% capability gap compared to US models, Zhou emphasized building a professional attack-and-defense team over relying on a single advanced AI. This approach aims to create a stable, continuously operating system, contrasting with what he described as the US focus on a "genius hacker" model. The company's strategy prioritizes developing a comprehensive operational capability rather than solely pursuing the most advanced individual model.

Source: Tech Radar