Who you are talking to
Lillith, the guide who remembers your story, is an AI. You are told so the first time you talk to her. The eight guides are real people. You can always ask who you are talking to, and you always get a straight answer.
Lillith the guide
Lillith is a conversational AI built on a large language model. She reads your birth chart, remembers what you share with her across conversations, and writes your daily reading. She is not a person and never claims to be one.
- The first time you open a conversation with her, the app tells you that she is an AI, before you write a word.
- Her messages carry her name, so you never confuse her with a guide.
- Her readings are generated for you from your chart and your words. They are guidance for reflection, not facts about the future.
The eight guides
Layla, Marc, Caprice, Sean, Nathan, Megan, Leo and Sophia are real people. When you open a session, a person reads your message and a person answers. Two things to know:
- Their messages may be translated by AI so that you can write in your language and they in theirs. The words are theirs; only the language changes.
- Notifications and daily readings are written by Lillith the AI from your chart, never by a guide, even when they mention one.
What is generated, and what is not
| In the app | Who or what |
|---|---|
| Conversations with Lillith | AI, disclosed at first interaction |
| Daily reading, month ahead, your chart | AI, computed from your birth details |
| The scan of a photo or a link | AI reading of what you send |
| Sessions with a guide | A real person, possibly translated by AI |
| Private notes and posts from a guide | Made by the guide |
| Notifications | Written by AI from your chart |
Ask, any time
If you are ever unsure, ask in the conversation: "Am I talking to a person or to an AI?" Lillith answers truthfully, and so do the guides. If anything in the app ever feels unclear on this point, tell us and we fix it.
What the AI does not do
- It does not diagnose, prescribe, or give medical, legal or financial advice.
- It does not decide anything about you. No score, no profile used against you, no automated decision with legal effect.
- It does not sell or share what you tell it. Your words are used to answer you, and for nothing else.
The models behind it
Lillith the guide, her memory, the reading of photos and the translation of the guides' messages run on Anthropic's Claude models. Voice messages you record are transcribed by OpenAI. Both are listed as processors in the privacy policy.
A question about how this works?
Contact support