How does content moderation work?

We want Questsmith to be a safe place for people to discover exciting content, while also freely exploring their own ideas and creativity in a private way.

Here is how that works:

• Unpublished single-player content is never moderated: We do not issue flags, suspensions, or bans for anything users do in private single-player play.
• Targeted AI boundaries: The AI is subject to boundaries designed to prevent it from generating unsafe content, specifically content that promotes or glorifies the sexual exploitation of children. There are no consequences to players if the AI hits these boundaries, and humans do not review those cases. Stories remain private.
• Seamless response filtering: We generate multiple possible responses for each user input so that if one option does not pass the filter, another can be delivered to the player.
• Transparent error handling: In rare cases, the AI may be unable to generate a response within its boundaries. When this happens, players see a message explaining what occurred and ways to continue their story.
• Story encryption: Stories are encrypted for extra player privacy. Stories are decrypted and sent to users devices when requested. The AI cannot process encrypted text, so decrypted story elements are passed to the AI to generate the next response.
• Published stories and Scenarios follow Community Guidelines: Unpublished content is never moderated. However, interactions with other users and content published in the Questsmith community are subject to our Community Guidelines.
• Scenarios go through an automatic rating check: Before publication, content is run through our rating checker. AI rates it as Everyone, Teen, Mature, or Unrated based on age appropriateness.

Even though AI may sometimes rate your Scenario to help enforce the rules, it is still important to understand the Community Guidelines before publishing. AI can make mistakes, but humans still provide the best judgement of what is right and wrong.

If you think the AI tool got it wrong, you can request a human review. Ping ata523100 on Discord for review.