AI Disclosure

Editorial articles in our palm-reading guides and blog are AI-drafted from a detailed editorial brief — one that specifies structure, tradition attribution, source citations, and the reflective/symbolic framing this site uses — then reviewed and edited by the ReadMyPalms team before publication. We do not claim these articles are human-written; we claim they are AI-drafted to a human-defined editorial standard and approved by a human before going live.

Our about pages, editorial standards, privacy policy, terms, and this disclosure page itself are written by the ReadMyPalms team. They describe the site's actual practices and are not generated on demand by a model.

The distinction we draw: an "AI-generated reading delivered to you as the product" (clearly AI; e.g. palm scanner, tarot interpretation, Ask Stella chat) versus an "AI-drafted article reviewed and approved by a human editor" (editorial guides, blog posts). Both involve AI; the difference is whether a human has shaped the final output before you see it.

We use a mix of large language model providers depending on the feature, latency budget, and what each model is best at:

• OpenAI (GPT family) — primary provider for many features
• Anthropic (Claude family)
• Google (Gemini family) — including vision models for image features
• Groq (serving Llama-based open-weight models, used in some lower-latency or higher-throughput cases)

The specific model handling any given reading may rotate based on availability, cost, and how the underlying models evolve over time. We update this list when the mix changes meaningfully.

It's a fair question. If palmistry and astrology are symbolic traditions designed to be practiced by a human, why hand them to a model? The honest reasons:

• ◆Modern LLMs have absorbed an enormous amount of palmistry, astrology, and tarot traditional knowledge from their training data — books, articles, manuals, forum discussions. They can speak the symbolic language of these traditions fluently.
• ◆They generate personalized readings 24/7 at near-zero marginal cost. That's what makes free, instant readings feasible at all.
• ◆They are not psychic. They are not predictive. They are not connected to any spiritual force.
• ◆What they actually are: sophisticated pattern-matching systems trained on traditional source material. They produce text that is statistically likely given the prompt and their training. That's the entire mechanism.

• ◆Treat readings as reflective prompts, not predictions. They're most useful when they help you think about something, not when they tell you what will happen.
• ◆Use them for entertainment and personal exploration. Don't use them as input to medical, legal, financial, or psychological decisions. If you're working through something serious, talk to a qualified human professional.
• ◆Remember you're conversing with software, not a human practitioner. There is no person on the other end of Stella's chat.

When you use an AI feature, certain inputs are sent to the AI provider serving that feature. Specifically:

• The photo you upload (for the palm scanner and aura scanner)
• The text you type (for the Ask Stella chat)
• Your zodiac sign or birth date when relevant to a reading

We do not send personally identifying information to AI providers unless you include it in your message. We do not train AI models on user data; we use the providers' APIs for inference only.

For full details on what we collect, how long we retain it, and how to request deletion, see our Privacy Policy.

If you have questions about how we use AI, want to flag a concern, or believe a specific reading went wrong in a way worth telling us about, email hello@readmypalms.com. We read every message, and we'd genuinely rather hear about a problem than not.

For the legal framing of what the readings are and aren't, see our Terms of Service. For data handling specifics, see our Privacy Policy. For the broader editorial frame, see our Editorial Standards and How We Research pages.

We'll keep this disclosure current as our use of AI evolves. If the model mix changes, or a new AI-powered feature launches, or our approach to a particular reading shifts, this page is where we'll say so.