Our structure
HASHTAG reduces common abuse surfaces: no profile images, limited room capacity, temporary content, request-based messaging and visible human management. Automated checks do not replace reports or accountable review.
Not allowed
Child sexual exploitation and abuse
Any sexual exploitation, grooming, sexualization, trafficking or endangerment of a minor is prohibited. Reports in this category are prioritized, preserved where legally required and escalated to appropriate authorities or designated reporting bodies.
Sexual exploitation and non-consensual content
No intimate imagery shared without consent, sextortion, sexual services involving coercion, or content facilitating exploitation.
Threats, violence and self-harm
No credible threats, glorification or coordination of serious violence, instructions intended to enable harm, or encouragement of suicide or self-injury. Supportive discussion and recovery-oriented conversation are permitted.
Hate and harassment
No attacks based on protected characteristics, dehumanization, targeted abuse, stalking, doxxing, unwanted sexual attention or coordinated harassment. Challenge ideas, never identities.
Fraud and dangerous deception
No scams, impersonation, phishing, malware, manipulated evidence intended to cause harm, or confidently presented high-risk falsehoods in health, elections, emergencies or public safety.
Illegal and regulated activity
No facilitation of trafficking, terrorism, illegal weapons, hard-drug sales or other serious illegal conduct. Contextual journalism, prevention and recovery discussion may be allowed.
Moderation actions
Depending on severity and confidence, we may block publication, reduce reach, add context, remove content, restrict messaging, suspend participation, preserve evidence or close an account. Child safety and credible imminent threats receive separate priority handling.
AI safety
#HASH output is clearly labelled and reportable. It has no higher privileges than a room’s founder or manager. Generative features must not create prohibited content or impersonate a human participant.
Reports, blocks and appeals
Reports should include the room, contribution and reason. Blocking takes effect immediately for your experience. We aim to notify affected users of material enforcement unless doing so would create additional risk or violate law.
Cooperation
We respond to valid legal requests and emergency disclosures under applicable law. We reject requests that are overbroad, invalid or inconsistent with fundamental rights.