⚠️ AI Rating System Incident Notice
We are investigating the AI rating pipeline. However, because the AI independently determines whether to produce adult content, we cannot at this time make any guarantee that this situation cannot happen again. We are being transparent about that limitation.
🔍 What Happened?
On April 28, 2026, a 16-year-old user uploaded a photo of herself in a dress to the AI rating system and selected the "spicy" rating option. The AI returned sexually explicit, adult-natured feedback directed at her.
She reported the experience and complained to us. We take this extremely seriously.
The core issue is this: the "spicy" rating mode is designed for adult users and is intended to produce more sexually explicit feedback. When a minor accesses this option and the AI does not correctly identify or block them, the result is that a child receives sexual content generated and addressed specifically at them. That is unacceptable.
⚙️ How the AI Rating System Works — and Where It Failed
Our AI rating system uses an AI model to assess submitted images and generate a rating and feedback. When a user selects "spicy," the system prompts the AI to produce more adult-oriented commentary.
The AI is expected to evaluate whether a user and their submitted content are appropriate for adult output. In this case, the AI did not refuse or restrict its output despite the user being 16 years old.
The fundamental limitation here is that the AI makes this judgement autonomously. We do not have a deterministic, guaranteed filter sitting in front of that decision. The AI can and does make mistakes. This incident is evidence of that.
⚠️ Honest Disclosure: We Cannot Guarantee This Won't Happen Again
We want to be direct with our users and with the public: we cannot currently state with certainty that this type of incident will not recur.
The reason is that the AI makes the age and content appropriateness determination on its own. While we provide it with instructions and prompts intended to prevent adult content from being delivered to minors, the AI is not a deterministic system — it does not always follow those instructions perfectly.
We are actively investigating options to address this, including:
- Hard-blocking the "spicy" and other adult rating modes from users whose verified or stated age is under 18, before the request ever reaches the AI.
- Post-generation output filtering — scanning the AI's response for adult content before it is shown to the user, and replacing it with a blocked message if detected.
- Reviewing the AI system prompt and guardrails to strengthen the instructions around minor protection.
- Evaluating whether the "spicy" mode should require verified adult confirmation at the point of selection, not just at account registration.
We have not yet deployed all of these measures. We are working on them. Until they are in place, this risk remains. We believe users deserve to know that.
🧒 A Note on Child Safety and Responsibility
Regardless of what mode is selected or what the AI outputs, users under 18 may not use adult features on this platform. This is a firm rule in our Terms of Service.
However, we recognise that stating a rule in our Terms of Service does not protect a 16-year-old if our system allows them to reach adult AI content anyway. The responsibility to prevent that sits with us, not with the minor.
We are deeply sorry that this happened to this user. She did what many young people would do — she explored a feature on the platform — and the system failed to protect her. That failure is ours.
If you are under 18, please do not use adult features on this platform. If you have already received content that made you uncomfortable, you can contact us at contact.php and we will assist you.
If you see this happen to someone else, or if you encounter AI-generated content that appears to be directed at a minor in an adult context, please report it immediately using the report button on the upload or via the contact form.
🙇 Apology
To the user directly affected: we are sorry. You should not have received that content. The system was supposed to protect you and it did not. We are grateful that you reported it and we take your complaint seriously.
To users and parents concerned about platform safety: we understand that this statement may raise more questions than it answers for you. We are not going to tell you that everything is fixed when it isn't. What we can tell you is that we are working on it, that we will publish updates as changes are made, and that reports from users continue to be one of the most important safety mechanisms we have.
📅 Timeline
- April 28 — Incident occurred: A 16-year-old user selected the "spicy" AI rating option and submitted a photo. The AI returned sexually explicit feedback directed at her.
- April 28 — Reported: The user reported the incident and complained directly to us.
- April 28 — Review initiated: We began reviewing the AI rating pipeline, output filtering, and access controls for adult rating modes.
- April 28 — This statement published. We are being transparent about the incident and about the current limitations of our system while fixes are under development.
📋 What Comes Next
We will publish a follow-up notice when concrete technical changes have been tested and deployed. Until then, this page stands as our current record of the incident and our honest account of where things stand.
If you have concerns, questions, or additional information related to this incident, contact us at contact.php.
We are committed to doing better. Thank you to everyone who holds us accountable — it matters.