Ritsora AI uses AI language models to help explain your financial information, summarise it, and suggest what may be worth reviewing. This page explains how that works and where its limits are.
1. What the AI can use
AI answers are built from the financial records in your own account and the figures Ritsora has calculated from them. Each figure passed to the model carries a label — Recorded, Calculated, Estimate or Scenario — and answers are expected to keep that label when quoting it.
2. Missing information is stated, not filled in
When the information needed to answer is not recorded, the honest answer is that there is not enough information yet, together with a list of what to add. Ritsora does not substitute a placeholder or a zero for a value you have not provided.
3. Wording safeguards
AI output is checked against a trust boundary in the product. That check flags figures that are not supported by your records, claims of live or real-time market data where no market source is connected, and guarantee-style wording such as “guaranteed”, “risk-free” or “best investment”.
4. AI can still be wrong
Language models can misread context, misinterpret a question, or reason poorly. Treat AI output as analysis to review, not as an instruction to follow. Always check the figures and assumptions shown alongside an answer.
5. Not personalised advice
AI-assisted insight in Ritsora is information and analysis. It is not investment, tax or legal advice, and it does not replace a qualified financial professional who knows your full circumstances.
6. Processing
To generate an answer, the relevant figures and your question are sent to an AI model provider for processing. We do not use your financial records to train third-party models, and AI usage is rate limited per account.