① Care should be taken, as generative AI outputs—particularly images or audio—may unintentionally infringe copyright. Before use, users should check for the existence of similar works (e.g. through image search). ② Users are encouraged to substantially modify or edit generative AI outputs before using them. 5) Transparency in the Use of Generative AI Where the use of generative AI may cause misunderstanding or confusion because audiences are not aware that generative AI was used, the resulting content shall clearly indicate that it was created with the assistance of generative AI. ① Where generative AI has played a substantive role in producing outputs—such as analyses generated with generative AI, or music, images, or videos created using generative AI—the work shall indicate that it was created using generative AI. ② Where generative AI is used to create outputs that may be confused with reality, such as deepfakes, this fact shall be clearly disclosed on the work. However, in the case of artistic or creative works, disclosure may be made in a manner that does not interfere with appreciation of the work. ③ In the case of generative AI systems that directly interact with external users—such as chatbots or real-time interpretation tools—users shall be clearly informed that they are interacting with an AI system. ④ This organization’s generative AI policy shall be made publicly available, for example through the organization’s website. 6) Consideration of the Environmental Impacts of AI As the use of generative AI expands, electricity and water consumption for operating data centers, as well as resource use for producing semiconductors for AI, continue to increase. Accordingly, generative AI should be used in ways that minimize negative environmental impacts. Generative AI Guide for Civil Society