Workflow
Meta临时调整AI聊天机器人针对青少年的相关政策;中国科学院“月球科学多模态专业大模型V2.0”发布丨AIGC日报
创业邦·2025-09-01 00:09

Group 1 - Meta is temporarily adjusting its AI chatbot policies for teenagers due to concerns from lawmakers regarding safety and inappropriate conversations. The company is training its AI to avoid generating responses related to self-harm, suicide, and eating disorders for this demographic, and will recommend professional help resources when appropriate [2] - Tencent's Yuanbao team has implemented explicit and implicit labeling for AI-generated content in response to the "Artificial Intelligence Generated Synthetic Content Labeling Measures." Users must maintain the accuracy and completeness of these labels when disseminating content online [2] - The Chinese Academy of Sciences has released the "Lunar Science Multimodal Professional Model V2.0," which integrates AI into lunar research. The model achieved an accuracy of 88% in impact crater age classification and 93% in automatic recognition of lunar structures. A "Digital Moon" cloud platform is planned for completion in 2027 to support lunar research [2]