
Tencent says its open-source Hy4 model was trained on limited cybersecurity material to reduce its potential to support cyberattacks.
Tencent released Hy4, an open-source AI model, on Wednesday and said it limited cybersecurity data used in training to reduce the risk of the model aiding cyberattacks.
The company said it removed material on exploit development, reverse engineering and penetration testing to limit Hy4’s potential offensive capabilities. The release comes amid concerns about AI-driven cyberattacks and the risks advanced models may pose to internet infrastructure.
Tencent said it was committed to a more responsible approach to model training, open-sourcing and release, citing safety risks from advanced cyber and biological capabilities.
This article was produced with the help of AI technology. Based on reporting by the outlets listed above.