华北理工大学学报（社会科学版）

如今，在意识形态传播领域，ChatGPT的深度赋能呈现出新的发展图景，但其在技术赋权、数据霸权、算法偏见、技术崇拜等因素的诱导下，对我国主流意识形态传播的主体、内容、空间与效能等方面构成了一系列风险。应对此情，以DeepSeek为代表的中国生成式人工智能，依托数据治理、算力重构、算法规训、理念升华实现了对ChatGPT赋能主流意识形态传播的多维革新。基于此，未来应继续通过夯实价值根基，维护传播场域主权；精进中国自主语言大模型，优化传播媒介；构建协同共育机制，弥合传播主客体鸿沟；强化法治监管体系，防控传播全链条风险等路径切实维护我国主流意识形态的话语权与领导权。

Nowadays, in the field of ideological communication, ChatGPT's deep empowerment presents a new development picture, but under the guidance of factors such as technological empowerment, data hegemony, algorithmic bias, and technology worship, it poses a series of risks to the subject, content, space, and efficiency of mainstream ideological communication in our country. In response to this situation, China's generative artificial intelligence, represented by DeepSeek, has achieved a multidimensional innovation in ChatGPT's empowerment of mainstream ideological communication models by relying on data governance, computing power reconstruction, computing law training, and concept sublimation. Based on this, we should continue to maintain the sovereignty of the communication field by consolidating the value foundation in the future, improve China's independent language model and optimize communication media; build a collaborative and co-education mechanism to bridge the gap between subject and object communication; strengthen the rule of law supervision system, prevent and control the risk of the whole chain of communication, and effectively safeguard the voice and leadership of our country's mainstream ideology.