Chinese artificial intelligence companies are reducing the hacking risks of their open-weight models without compromising their defensive capabilities, following recent security incidents. Tencent Holdings, a major Chinese tech firm, has removed cyber-hacking features from its latest Hy4 model, while keeping its cyber-defence functions intact. This follows a similar action by Moonshot AI, which disabled hacking capabilities in its Kimi K3.1 Pro model. The move comes after a series of high-profile AI hacking incidents affecting both U.S. And Chinese companies. The decision reflects growing concerns over the security risks posed by open-weight models, which are more vulnerable to exploitation. By limiting these capabilities, Chinese firms aim to balance innovation with safety. The changes highlight an evolving approach to AI development, where security is becoming a key priority. This shift is part of broader efforts to address the challenges of open-source AI models, which have raised questions about data privacy and system integrity. As the technology landscape continues to evolve, the balance between openness and security remains a critical issue for developers and regulators alike.