VRChat’s developers said on October 28 that they will not add labels for AI-generated content or filters to hide it. The team is concerned about inaccurate labeling and harassment of creators accused of using AI.
The survey received 21,595 responses in a week. In results weighted toward VRChat’s player base, 95% of respondents wanted some form of AI-use labeling. Responses across eight example categories produced 3,141 different labeling combinations, with none supported by more than 10% of respondents.
The team aims to add positive and negative world ratings by the end of 2026 or early 2027. Multi-image world galleries are targeted for the first half of 2027. An optional mark for human-authored worlds and a field listing tools used remain under consideration.
