Acceptable Use & AI Content Policy

Renderball is a business-to-business design tool. Customers use it to make presentations, social posts, one-pagers and similar business designs: pages made of text, charts, diagrams, tables, shapes and layout. Output is editable document content — every element remains a real object the customer can move, rewrite or delete — not a flattened picture.

Renderball uses third-party language models to do three things: draft an plan from the customer’s own brief, write the layout and copy for each page as editable document code, and regenerate a single element on request. A vision model checks rendered pages for legibility problems such as low contrast or overlapping text.

Optionally, and only when the customer supplies their own website address, we read that page to extract their brand’s colors, fonts and logo so the document matches their brand.

Renderball does not generate photorealistic images of people, human likenesses, faces, voices, or video of any person. It has no face-swap, voice-cloning, avatar, or “deepfake” capability and no feature that edits a photograph of a person. It does not generate adult content. Where a document needs a photograph, Renderball uses licensed stock imagery from Pexels or an image the customer uploads themselves.

Text a customer provides — their brief, their document copy, their brand guidelines — is sent to our model provider, Anthropic, to produce the document, and to no other model provider. Rendered page images are sent to the same provider to check the finished pages. When a customer connects their own AI assistant, what they send it is governed by that assistant’s own terms. We do not sell customer data, and we do not use customer content to train models. Our full vendor list is in the Privacy Policy.

Generated documents are checked before a customer sees them: automated gates verify that every factual claim in the output traces back to something the customer wrote or to text read from their own website, so the model cannot invent statistics about their business. Generated document code is executed in an isolated process that holds no credentials and is stopped automatically if it does not finish. Uploaded files are type-checked by content rather than filename, and uploaded vector images are stripped of any embedded script.

You may not use the Service to create or distribute:

Content that infringes anyone’s intellectual property, including brand assets you are not authorised to use. Content that impersonates a real person, company or public authority. Fraudulent or deceptive material, including fake invoices, receipts, credentials, or financial documents presented as genuine. Content that harasses, defames or discriminates. Content that is unlawful in the jurisdictions where you or your audience are located. Malware, or any attempt to make the Service execute code outside the documents it produces.

You are responsible for the documents you make and for having the rights to any brand assets, imagery, fonts or copy you supply.

Report misuse to [email protected]. We investigate every report and may suspend or terminate accounts that breach this policy. Where we are legally required to act on a report, we will.