Recently, the release of ByteDance's Seedance 2.0 video generation large model has sparked heated discussions. Tim (Pan Tianhong), the founder of self-media film and television hurricane, commented, "I didn't give any prompts, any words, or any information, and I didn't give my voice file. I just uploaded my face, and this AI actually knew the voice of this face." Tim said that he personally did not give official authorization, nor did he receive any money, which made him feel horrified. In this case, without uploading the corresponding material, the model generated highly similar sounds and picture styles, which also attracted widespread attention.
At the same time that related events were brewing, ByteDance Seedance 2.0 urgently suspended its real-person material reference capabilities. An official operator said: Hello everyone, Dream Creators, Seedance 2.0 has received much more attention than expected during the closed beta period. Thank you for your use and feedback. In order to ensure the health and sustainability of the creative environment, we are making urgent optimizations based on feedback. Currently, we do not support the input of real-life materials as the main reference. We know that the boundaries of creativity are respect. We apologize for the inconvenience caused to you during the product adjustment period. We look forward to officially meeting you with a more perfect appearance.
