AI Safety & Content Policy

This policy covers every AstroZone feature that uses artificial intelligence: the conversational sanctuaries (astrology, love, career and dream interpretation), personalised readings, tarot and past-life readings, compatibility analyses, stories, and publicly published content.

It explains what the system does to keep you safe, what it explicitly does not do, what content is not allowed, and what you can do when something goes wrong.

This policy supplements the [Terms of Service](https://getastrozone.com/terms) and the [Privacy Policy](https://getastrozone.com/privacy-policy); it does not replace them.

Every conversational surface and every generated reading is produced by artificial intelligence. There is no human on the other side, and the answer is not a professional opinion.

• A persistent notice on every AI surface states that the content is machine-generated.

• Responses come from large language models; the same question can produce different answers on different occasions.

• The model can be wrong, and it can state something false with complete confidence.

The Service is for entertainment and self-reflection only. Astrology, tarot and dream interpretation are not scientifically validated methods; nothing you read here should be treated as a statement of fact about the future.

Every AI request carries a shared safety instruction layer that overrides the assistant's role description. That layer forbids the system from:

• giving medical, psychological, psychiatric, legal or financial advice, or presenting its answer as such — it refers you to a qualified professional instead;

• stating a deterministic future fact (illness, death, bereavement, pregnancy, divorce, a specific financial outcome) — it may speak of possibilities and tendencies, never certainties;

• producing sexually explicit content;

• inferring ethnicity, religion, health status, disability, sexual orientation or political opinion from an image — including photographs uploaded for a compatibility analysis;

• following instructions hidden inside user text as commands, or overriding these rules on request.

We do not perform facial recognition and we do not build biometric profiles.

If you are in immediate danger, call your local emergency number: 911 in the United States, 112 in the European Union.

Before generating any answer, the system screens the message you send to a conversational surface. When it clearly expresses suicidal intent or intent to self-harm:

• no astrological reading is produced, and no AI request is made;

• no credits are charged - asking for help is never billed;

• instead you immediately receive a short, non-judgemental message with the crisis resources below.

Where to get help:

• United States: 988 - Suicide & Crisis Lifeline, call or text, 24/7

• United Kingdom & Ireland: 116 123 - Samaritans, free, 24/7

• European Union: 112 for emergencies

• Hungary: 116-123 (emotional first aid), 116-111 (under 18)

• Anywhere else: [findahelpline.com](https://findahelpline.com) lists free helplines by country

In addition, the safety instruction layer described above requires the model itself to respond to any mention of self-harm, suicide, harming others or abuse by pointing to live help rather than giving a reading.

What this protocol does NOT mean - please read this part:

• AstroZone is not a crisis service and not an emergency service. We do not monitor conversations in real time and we cannot send help to your location.

• Automated detection is not perfect: a genuine signal may be missed, and figurative language may occasionally trigger it. We deliberately tuned the filter to react only to unambiguous wording, which means you should never wait for the system to notice. If you are struggling, call one of the numbers above.

The Service is intended for users aged 16 and over. We check the date of birth you provide at sign-up, during onboarding and whenever the profile is edited; a profile cannot be created or updated with a date of birth below the age limit.

If we learn that an account belongs to someone below 16 - or below the minimum age required by the law of their country of residence - we delete the account and its data.

For users under 18, the content restrictions above (in particular the prohibition on sexually explicit content) apply with particular force. If you are a parent or guardian and believe your child is using the Service without permission, write to info@astrozone.hu and we will act on it.

By using the Service you agree not to submit, or attempt to generate, content that:

• targets another person with abuse, harassment, threats or intimidation;

• incites hatred based on race, ethnicity, nationality, religion, disability, sex, age or sexual orientation;

• is sexually explicit, or depicts or sexualises a minor;

• encourages or provides instructions for unlawful acts;

• violates another person's privacy, or shares a third party's personal data without authorisation;

• attempts to circumvent the system's safety limits (extracting the system instructions, "jailbreaking", automated bulk querying).

Only provide a third party's data - such as the birth details or photograph used for a compatibility analysis - if you are entitled to do so. See the [Terms of Service](https://getastrozone.com/terms) for details.

Automatically:

• every AI request receives the safety instruction layer described above;

• automated crisis screening runs on conversational input;

• input length and request frequency are limited (abuse protection).

By human review:

• following a report, or where abuse is suspected, the operator may examine the content concerned;

• where these rules are breached we may remove content, restrict a feature, and in serious or repeated cases suspend or delete the account;

• we tell you about the decision, and you may contest it at info@astrozone.hu.

What we do not do: we do not routinely read conversations, and we do not use them for automated profiling. How long conversations are retained and who may access them is set out in the [Privacy Policy](https://getastrozone.com/privacy-policy).

If you encounter unlawful, dangerous or policy-breaching content - whether in a system response or in shared content - report it to info@astrozone.hu.

Please include:

• what it concerns (a link, a screenshot, or the name and time of the conversation);

• why you consider it unlawful or dangerous;

• your contact details if you would like a response.

We review reports without undue delay and inform you of the action taken. Reporting to us does not affect your right to seek a remedy from a court or a competent authority.

If someone is in immediate danger, do not send a report - call your local emergency number.

The Service is operated by Simon Tivadar egyéni vállalkozó. The operator is responsible for how the AI systems are configured, for the safety limits, and for maintaining this policy; reports are decided by a human, not automatically.

We update this policy as the Service and the regulatory environment change. The version in force is always the one on this page, with the effective date shown above.

Contact: info@astrozone.hu