How Signal handles your research.

Research questions and OpenAI

When you submit a question, it is sent to the Yellow Bird Tech hosting service and then to the OpenAI API. Signal uses OpenAI’s GPT‑5.5 model and web search to prepare an answer with source links. Your question may be used as a web-search query or to select public pages relevant to the answer.

Signal requests that OpenAI not store the response as application state. OpenAI states that API data is not used to train its models unless the API customer opts in. OpenAI may retain prompts, responses, and related metadata in abuse-monitoring logs for up to 30 days under its default API data controls.

Please do not enter private, confidential, medical, financial, or other sensitive personal information into the research box.

Information kept by this site

• Rate limits: the server uses a one-way hash of your IP address with temporary timestamps to enforce hourly and daily request limits.
• Library and watchlist: saved research, recent questions, followed artists, and theme preferences are stored in your browser on your device. They are not synced to an account.
• Feedback: a helpful or not-helpful vote stores the vote, an internal research reference, and the time received. It does not store your question or answer.

Web sources and links

Signal searches public web sources and displays links to third-party sites. Those sites have their own privacy practices. Review the cited sources before relying on a research result.

Your choices

You can use the site without saving research or following artists. Clear this site’s browser data to remove its locally saved library, watchlist, and preferences.

For more about OpenAI’s API practices, see OpenAI’s data controls. For questions about this beta, visit Yellow Bird Tech.