AI and your data - what the AI sees

Overview

Some features use an AI model: the copilot in the app and on Telegram, explanations of metrics, and short summaries of unusual options activity. This article states exactly what those features send to an AI provider and what they never send. The same text is on the public page What the AI sees.

Which providers

OptionsScreenerPro uses the free tiers of three model providers, in this order, falling back to the next when one is unavailable: Groq, Cerebras and Google Gemini. An administrator can turn any of them off. When none is configured, AI features say so plainly and the rest of the product works as usual.

What is sent

• The question you type, and the copilot's earlier turns in the same conversation.
• The minimum context of the page you are on: the page name, the symbol and expiry you are looking at, the screen filters, and the rows the answer is about. Market numbers are sent as they appear on screen, labelled delayed.
• For help questions: the text of the help articles that match.

What is never sent

• Your password, payment details, API keys or session tokens.
• Your email address, name or IP address.
• Other users' data.

How long it is kept

AI request records (prompt, response, tokens, timing) are kept for 30 days to investigate problems and control cost, then deleted automatically. Copilot and Telegram conversation turns are kept for 90 days. Providers' own retention is set by their terms; OptionsScreenerPro uses settings that do not allow training on your data where the provider offers that choice.

Your choices

• Do not use the copilot: nothing is sent.
• An administrator can switch AI off for everyone with a kill switch.
• Ask support to delete your AI records sooner.

Troubleshooting

The copilot says AI is unavailable

No provider is configured, the monthly AI budget is used up, or the providers are down. Everything else keeps working; try again later.

Related

• FAQ
• Understanding data freshness
• Metric glossary