AI Content & Reporting
Grape generates text with AI. If something it produced is harmful, offensive, or plainly wrong, tell us here and a human will look at it.
Grape uses generative AI
Chat, summaries, flashcards, quizzes and mind maps are produced by large language models. Model output can be wrong, out of date, biased, or made up entirely, and it is not legal, medical or financial advice. Check anything important against a source you trust before you rely on it.
Where your input goes
When you use an AI feature, the note content or question you send is passed to the model provider handling that request. Depending on the model you pick that is OpenAI, Anthropic, Google, Groq, xAI, DeepSeek or OpenRouter. We do not use your content to train models, and we do not store AI conversations on our servers. If you bring your own API key, the request goes straight to that provider under their terms.
Reporting from inside the app
Every AI answer in the app has a report button underneath it. Tapping it sends us the answer along with your note, so you do not have to copy anything by hand.
What we do with a report
Reports go to a human. We review each one, and where the problem comes from how Grape prompts or configures a model we fix it. Where it comes from the underlying model we pass it to that provider and, if the output is severe enough, we drop the model from Grape.