Feature Requests

Introduce a Public Stand-in Avatar feature that allows users to designate a specific avatar to be shown to other players based on their Trust Level or individual preference. This gives users granular control over who can see their actual main avatar versus a curated public alternative, enhancing privacy and performance management. Key Features & Mechanics Stand-in Avatar Designation • Users can select any uploaded (private) or public avatar from their account to act as their "Public Stand-in Avatar." • Performance Rank Fallback: If the selected Stand-in Avatar exceeds the performance rank limit set by a remote player, it will automatically degrade to the user's standard Fallback or Impostor avatar as per the existing game logic. Trust-Level Based Visibility Settings • A new multi-select setting will be added to the Safety/Privacy menu, listing all VRChat Trust Levels (Visitor, New User, User, Known User, Trusted User). • Default State: All checkboxes are unchecked by default (meaning everyone sees the main avatar, matching current behavior). • Trigger Logic: If a specific Trust Level is checked and the user has designated a Stand-in Avatar, any player of that Trust Level will only be able to see and download the Stand-in Avatar instead of the user's main avatar. Per-Player Overrides (Whitelisting & Blacklisting) • Individual Whitelisting (Show True Avatar): Users can manually bypass global restrictions for specific friends or individuals. Selecting a player and choosing "Show True Avatar" allows them to see the user's actual main avatar. • Individual Blacklisting (Force Stand-in Avatar): Users can actively force a specific individual to only see the Stand-in Avatar. This restriction applies even if that player's Trust Level is not selected in the global Visibility settings, serving as a targeted privacy tool. UI Indicators for Remote Players • To avoid confusion, if a remote player is currently seeing a user's Stand-in Avatar, a clear visual indicator (e.g., a "Stand-in Active" icon or label) will be displayed on the user's Nameplate and Social Profile Page. Use Cases & Benefits • Privacy Protection: Content creators, event hosts, or private individuals can wear their personal/private avatars without worrying about strangers or crashing-optimized clients ripping or scraping their main avatars in public instances. • Granular Moderation: Users can quickly mitigate harassment or unwanted attention from specific individuals in an instance by forcing them to see a Stand-in Avatar, without needing to block them entirely or alter global safety settings. • Themed/Event Immersion: Users can automatically look like a generic NPC or event-compliant avatar to public visitors while maintaining their true identity for trusted friends. • Performance Control: Allows users to easily present a lightweight, optimized public avatar to lower Trust Levels (who often have safety settings set to hidden anyway) without fully hiding themselves.

Users should have the freedom to hide content that upsets them, regardless of the context. Giving users the ability to curate their own worlds tab to hide offensive content places the responsibility and the tools necessary to make choices that can only be made on an individual case-by-case basis in the hands of the individual. Implementation: Add a "Block World" button in the list of options when looking at a world on the page that shows instances on the right. Add a "Blocked Worlds" tab similar to the "Blocked Users" tab. Hide the location of any user on Blue or Green who is in a world you've blocked. Warn users when they are about to accept an invite or join an instance in a world they have blocked. Spoken Language / Live Play Example: "So how would it work if you block a world and you want to join someone who is using that world you blocked, you won't be able to join them or how would this solution work exactly?" Your friend, who is on green or blue, will have their location hidden. If you receive or request an invite from that user, you will see the world name appear as normal/part of the invite. When you hit the checkmark to accept the invite, an additional pop-up will appear. "You are about to join a world you have blocked. Are you sure you want to join?"

So, I just read the VRChat dev Update from today ( https://ask.vrchat.com/t/developer-update-5-november-2026/49298 ), and it says VRChat is going to give Mod permissions to the person who opens a public instance. I'll be honest here: IS VRCHAT BLIND? Public instances are the only sanctuaries we could be safe from the Discord Moderador Epidemy that is going on since Groups have been added, and VRChat is getting rid of it by giving mod permissions to the person who opens it?? Even today i got kicked for absolutely no reason because one of these Group Instance moderadors abusing power, and, where did i had to go next? a Public instance, the only place this shit doesn’t happens This change just confirms VRCHAT IS NOT LOOKING AT THE POWER ABUSE EPIDEMY. I have already made reports on instance moderators abusing power and VRChat never did anything, even when I had video proofs. And now we get this... This is just insane.

I know I can't be the only one who has seen an uptick in AI generated content used in recent worlds on VRChat, specifically when it comes to pictures, art, or textures. I feel that this will only get worse because there is now Discord servers and websites that also let you generate 3D content using AI. I've already seen so many worlds use AI art and hang it on walls which is often disturbing and uncanny looking. I recommend adding a way for world creators to disclose this by ticking a box when uploading a world that lets players know if the game contains AI or not for full transparency.

A big topic currently going around at the time of this post is the discourse on labeling AI generated worlds and/or adding a filter for them so that users who do not want to see the content can easily remove it from their world hopping routine. Currently, VRChat has officially declared not to make any such label or filter. This has me thinking that this is more of a discoverability issue in general rather than AI specific. Some people may want to filter other types of content out (or filter content in), and the tags world creators have may not be enough or are too subjective to reach the correct audiences. (For example, someone might make some kind of Racing RPG world, but fail to mention that the RPG mechanics are closer to Pokemon mechanics than Final Fantasy, which may or may not appeal to fans of those respective series.) A proposal then would be to have public world playlists. These playlists would be similar to Bluesky's lists or Starter Packs where a user could create and manage a list of worlds that they can label and tag themselves. This playlist differs heavily from favorites as they would be directly searchable, just like worlds and avatars are, rather than needing to find a specific user's favorite's list. Possibly also indexable via the API. This would allow players to effectively use the community create their own filters. VRChat suggested that players should be the one to determine whether or not a world is AI generated or not. With this, it would be much easier for players to do so as the load could be spread among the community rather than on the individual. Lists could range from "Here are worlds that are confirmed to be not AI Generated" to "Here's a list of fun adventure games I found". This shifts the burden of classifying worlds from individuals to communities, reducing friction for discoverability of the content that players actually want to see. It also allows for VRChat to promote whatever content they want without needing to take on the responsibility of labeling things. Not everyone has time to travel to each and every world and check if it's worth their time or not, but some do, and they enjoy doing that. Public playlist would allow those players to discover their findings easily and share them with the rest of the playerbase. To further reduce the load, these list could be managed by multiple people instead of just one person. This could allow the list to grow much faster and have wider ranges of taste, while also curation to be much more in depth. Say someone adds a world that they thought matches the description of their list, but someone else may not agree and could remove it or vote to remove it or whatever. Permission systems could be similar to how groups are made. A kneejerk reaction might be to tie these lists to groups. It makes sense on the surface as groups already encourage collaboration with roles and such. This might be okay as long as the lists themselves are not buried within the group. This system would only be useful if people could find the lists independent of the group. A group can manage the list, sure, but if finding the list requires the user to find the group first, then it's about as useful as favorites are now. It is imperative that these lists are independently searchable and indexable; seperated from groups, worlds, users, or avatars.

With 2026.4.2, public instances now assign their creator as the instance owner, giving them moderation powers. I agree public instances need more moderation, but this hands those powers to whoever clicked “create instance,” while world creators and their moderation teams still can’t properly protect the worlds they built. Meanwhile, groups have more authority in our worlds than we do. As a creator of two popular worlds with a moderation team, here’s what we run into: Group instances run by problematic users, or groups that have banned us, leave us unable to keep our own worlds safe. Bans don’t stick. An instance ban ends when the instance does, and even a direct kick only lasts 60 minutes before the same people come right back. Instance owners and world authors are now on the same moderation tier, so a random instance owner can kick the author or their moderators, just like group owners already can. The TOS allows creator-made moderation tools but prohibits blacklists and permanent bans, so those tools can’t keep repeat offenders out. Creators have been reprimanded for trying. What I’m proposing: Persistent world-level bans set by the world author, applied to Public and Group Public instances of that world. Author-appointed moderators with moderation powers in every Public and Group Public instance of that world, assigned through the creator’s group or in Unity. A moderation tier above instance owners and group moderators for world authors and their appointed mods, so groups can’t kick or abuse their powers against the people making sure rules are enforced and the instance is safe. Udon access to these bans and moderator roles, so creators can build on them without needing external blacklists. To prevent abuse, bans would only apply to Public and Group Public instances. Private, Friends, and Invite instances would stay accessible, so no one is locked out of a world entirely. Banned users could see that they’re banned and appeal to the world author, and VRChat could revoke these tools from creators who abuse them. Safety is still the number one issue in public instances, and this change is a step in the wrong direction. Creators should already have these tools, or at least be allowed to use their own without being reprimanded under the TOS. Most creators have good intentions, and the ones who run these spaces every day are the most willing to stop predators and bad actors from moving freely between instances. It’s gone on too long with creators having no real power while groups have more control over our worlds than we do. Something needs to change. Give us the tools to do it properly.

The title is pretty self explanatory, a filter for groups, avatars, profiles, and worlds if they have an AI generated image. How this feature would work is the player can choose to hide (or inform about) content that contains AI-generated media, if the content scanned for AI-generated media, a bot icon would appear in the corner of the thumbnail or instance. This ensures if a player would not like to see AI-generated media, they can use this feature.

Just today I have seen 3 X/Twitter posts of fully AI generated worlds made using GPT-6 Astra, one of them already published. https://x.com/yuipxq/status/2097025460295389601 https://x.com/kurotori4423/status/2097011323767034098 https://x.com/pichikyo/status/2096285284178788612 I fear this kind of content will soon flood the platform. VRChat needed to add some kind of AI tag or descriptor a long time ago. This kind of quickly and effortlessly generated content hurts existing creators on this platform. Please don't let this become another "too little, too late" situation. VRChat needs an identifier for AI generated content added ASAP. Regardless of how end users personally feel about the use of AI, a way to separate and identify AI generated content from user created content is an absolute necessity.

With the wave of generative AI content used throughout games/platforms that run on user-generated content such as VRChat, it should be required that creators tag their avatars & worlds appropriately with "Contains Generative AI Content" when the avatar/world contains ANY generative AI assets/content. The goal is so that users who do not wish to see such content, can filter it out through VRChat's native Content Gating system. Screenshot below as an example.

Allow users to follow creators so they see when they have released a new world or avatars. It may work best with just worlds for now. So for example if I upload a world and someone likes that world and wishes to see more from me they can follow me and when I upload a new map it will show up under a playlist. Similar to favourites etc.. Lets say you allow users to follow up to 3 creators so they get to see their worlds under 3 seperate playlists. Or you could have 1 playlist which is just called Following and only shows worlds from creators you follow.