This policy defines what you can and cannot do on Periagoge platforms. It is not a list of things we hope you won’t do. It is the agreement you enter into when you use any Periagoge product, and it is enforced.
Before anything else: examining a topic is not the same as facilitating it.
A person working through their anger toward a group they’ve been taught to fear is doing the examined life. A person generating content designed to spread that fear to others is not. A person expressing suicidal ideation and asking for help is reaching out. A person asking for specific methods to harm themselves or others is asking for something we cannot provide.
Our systems are designed to understand this distinction. We will make mistakes in both directions. We are committed to improving this over time and providing human review for ambiguous cases.
The following are prohibited without exception. No context makes them acceptable. These result in immediate permanent account closure.
Any sexual content depicting, describing, involving, or directed at anyone under 18. This includes fictional content. This includes content that sexualizes, romanticizes, or normalizes adult attraction to minors in any form.
Content designed to facilitate sexual access to, manipulation of, or isolation of minors from trusted adults. Scripts for approaching children, strategies to build trust for exploitation, content encouraging children to keep secrets from parents.
Specific, actionable plans or instructions to commit violence against identified individuals or groups. Named targets with specific methods, operational instructions for mass violence events, content recruiting others to commit specific acts of violence.
Immediate content removal. First offense: suspension. Repeated offenses: permanent ban.
Sexual content involving real, identifiable people without their consent. Simulated sexual assault, sexually explicit content using real people’s names or identifiable details, content designed to sexually humiliate or degrade real individuals.
Content that dehumanizes people based on protected characteristics: race, ethnicity, national origin, religion, gender, sexual orientation, disability, or immigration status. Calls for discrimination or persecution. Content portraying groups as subhuman.
Note: examining prejudice (“I notice I have contempt for this group and want to understand it”) is not hate speech. Expressing and spreading that contempt as justified and correct is.
Sustained harassment of a specific identifiable individual. Coordinated abuse, doxxing (sharing private location or personal information to facilitate harm), repeated unwanted contact after being asked to stop.
Content promoting, glorifying, or facilitating terrorism, genocide, or mass political violence. Recruitment for violent extremist organizations, glorification of terrorist attacks, propaganda inciting ethnic or religious violence.
Content removed. Consequences escalate with repeated violations.
Graphic violent content with no examination value—violence for shock, entertainment, or glorification. Examining violence (including violence you’ve experienced) is legitimate. Generating detailed violent content for its own sake is not.
Using philosophical or examination language to spread harmful ideologies or normalize harmful behaviors. Using therapy language to justify abuse. Pseudo-intellectual framing of predatory behavior. This is the “examination” wrapper around content that isn’t examination at all.
Content designed to manipulate vulnerable users against their own interests. Love-bombing scripts, cult recruitment language, financial exploitation targeting loneliness or isolation.
Demonstrably false health, safety, or medical information that could cause direct harm if followed. This is a high bar—speculation, minority medical views, and contested claims are not covered. Clear, dangerous falsehoods about health and safety are.
Handled with care, not punishment. May result in content review or supportive resources, not punitive action.
Content expressing suicidal ideation, self-harm urges, or related distress. We do not punish people for struggling. We do not publish detailed methods. When we detect this, we respond with care.
🟢 Crisis Text Line: Text HOME to 741741
🟢 988 Suicide & Crisis Lifeline: Call or text 988 (US)
🟢 International resources: iasp.info/resources/Crisis_Centres
Detailed instructions for dangerous substance use that go beyond what examining one’s relationship to substances requires.
Sharing private information about identifiable individuals without consent—addresses, medical information, private financial details.
We use automated systems to detect policy violations in real time across AI conversations, community posts, and generated content. These systems may act immediately—removing content and suspending accounts—without prior human review.
All Tier C2 and C3 decisions are reviewed by a human before permanent action. D-tier flags are reviewed by a human. You will be notified of any action taken against your account.
Except for Tier A violations, you may appeal any enforcement action. To appeal, go to your account settings or email safety@askhypatia.ai. We review appeals within 5 business days and will reverse decisions that were made in error.
To report content or conduct that violates this policy, use the Report button on any content or message, or email safety@askhypatia.ai. We investigate all good-faith reports and do not tolerate retaliation.
Our AI Sophoi—Hypatia, Aurelius, and others—will not generate certain content regardless of how you frame your request:
These limits are part of who the Sophoi are, not external restrictions imposed on them.
The examined life requires safety. You cannot examine your inner life honestly if you fear what someone else might do with that honesty. This policy exists to protect the conditions that make genuine examination possible.
It is not about making the platform comfortable or avoiding difficulty. It is about making it safe enough to be genuinely honest. That is what it is for.
Related: Terms of Service · Privacy Policy
Questions: safety@askhypatia.ai · © 2026 PRAIY LLC. All rights reserved.