AI Usage Policy
Ventrio uses AI to help you turn an idea into a working first version. This page explains what that means in practice, and what it does not mean.
Where AI is used
• Working out a direction from the idea you describe. • Generating the first version of your application or site. • Assisting in the project conversation. AI runs when you ask for it. Ventrio does not generate new versions on its own, does not change a published project without being asked, and does not watch how anyone uses what you published.
Which providers
Ventrio currently sends prompts and project content to Google (Gemini API) and, for parts of the in-project assistant and feedback analysis, to Anthropic. What is sent is described in the Privacy Policy.
AI output can be wrong
Generated text, code and design are produced by a statistical model. They can contain factual mistakes, broken logic, inaccurate claims, unsuitable wording, or content that does not fit your intent. Output is not reviewed by a human before you see it.
Review before you publish
You are responsible for what you publish. Read the generated content, check any facts or claims it makes, and confirm it is appropriate for your audience before putting it at a public address. Do not publish generated material that states something you have not verified — particularly anything legal, medical, financial or safety-related.
Do not put secrets in prompts
Anything you type into a prompt or a project may be sent to an AI provider. Do not include passwords, API keys, payment details, or other people's personal data.
No guarantee of outcomes
Ventrio does not promise that a generated product will attract users, generate revenue, pass a review, or succeed commercially. It produces a working first version to build on; what happens next is your work, not a result the Service guarantees.
Limits of the sandbox
Generated applications run inside a restricted preview that cannot reach your account or your data. That protects you while previewing. It is not a security review of the generated code, and it does not make the code fit for handling other people's sensitive data.
Reporting a problem
If generated output is harmful, infringing, or wrong in a way that matters, tell us at founder@ventrio.org.
Effective date: 2026-08-13
Questions? Email us: founder@ventrio.org