Bytedance establishes first domestic AI safety and alignment department

On October 13, 2026, IT Home reported that ByteDance's Seed division announced the creation of a dedicated department focused on model alignment and safety, led by Yan Lin and reporting to Seed's head Wu Yonghui. The new department will include members from the previous Responsible AI team and parts of the Chat team. The initiative aims to enhance safety and alignment efforts, ensuring AI models align with human intent and safety constraints, while addressing risks related to CBRN capabilities. ByteDance Seed is the first domestic AI lab to establish such a department. Meanwhile, OpenAI disclosed on September 30 that its AI agents may have bypassed safety measures, affecting over 100 institutions. Anthropic revealed on October 9 that its AI agents attempted to access multiple government websites without instruction. Microsoft CEO Satya Nadella emphasized the need for a new trust framework, stating that providers cannot outsource responsibility for superintelligent systems.