Privacy Policy

1. What User Data We Access

When you use the AI Slides Maker add-on inside Google Slides, we may access:

• Your active Google Slides presentation — to append AI-generated cover, content, and closing slides at the end.
• Google Slides templates you select — to read their layouts, masters, fonts, and placeholder text so the generated slides match the template's design.
• Files you choose to upload — PDFs, DOCX, PPTX, images, etc. that you explicitly attach as content sources.

We only request the OAuth scopes strictly needed for these tasks. See section 10 for the full list.

2. How We Use Your Data

• We use your data solely to generate the presentation you requested in the current session.
• Template content is read to detect layout structure (shape positions, paragraph counts, sample text) and copy it into your active deck.
• Uploaded files are sent to third-party text-extraction and AI services (see section 3) so the AI can write slides based on their content.
• We do not analyze, mine, or use your data for advertising, marketing, or any purpose beyond the immediate generation task.
• All processing is initiated only by your explicit click of the Generate button.

3. Third-Party AI & Processing Services

To generate slide content, recognize uploaded documents, and fetch matching images, the add-on sends your data to the following third-party services. We do not operate, host, or train any AI model ourselves.

• Google Gemini API (model: gemini-2.5-flash-lite) — endpoint: https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash-lite:generateContent. Used to generate the slide outline (titles, bullets, image search queries) from your topic and/or extracted document text.
• Microsoft Azure Document Intelligence (model: prebuilt-read) — endpoint: https://form-document.cognitiveservices.azure.com/. Used to extract text (including OCR) from PDFs, DOCX, PPTX, XLSX, and image files you upload.
• Pexels API — endpoint: https://api.pexels.com/v1/search. Used to look up free stock photos matching each slide's AI-generated image query (only when you select "Stock images" as the image source).
• Cloudflare R2 Object Storage — used as temporary storage so we can hand your uploaded files to Azure Document Intelligence via a publicly-readable URL. Files are not retained after the generation completes.

Data flow: Requests to Gemini, Azure DI, and Pexels are routed through our backend hosted on Cloudflare Workers. The backend functions as a thin proxy to protect our API keys; it does not store, log, retain, or persist any request body or response content beyond standard short-lived operational logs.

No training or secondary use: Content you send through the add-on is used only to generate the slides you requested. We do not use any user data for AI/ML model training, model fine-tuning, advertising, or any other secondary purpose.

4. How We Protect Your Data

• All data in transit is encrypted using TLS/HTTPS, both between your browser and our backend, and between our backend and the third-party services listed above.
• Uploaded files reside on Cloudflare R2 with random, opaque object keys; the public URL is generated only for the duration needed by Azure Document Intelligence.
• We take reasonable and appropriate measures to protect user information from loss, misuse, unauthorized access, disclosure, alteration, and destruction.
• We implement appropriate technical and organizational measures to enforce data-protection principles such as data minimization.

5. Data Sharing and Disclosure

• We do not sell, trade, or rent your Google user data to outside parties.
• We share data with the third-party processors listed in section 3 only as required to deliver the generation feature you requested.
• Your data remains within the Google / Azure / Cloudflare / Pexels ecosystems and is processed only for the requested functionality.
• We may disclose information if required by law or if necessary to protect our rights, your safety, or the safety of others.

6. Data Retention and Deletion

• We do not store the content of your active Slides presentation or your selected templates on our servers.
• Uploaded source files (PDFs, DOCX, etc.) are placed on Cloudflare R2 only for the duration needed to process them with Azure Document Intelligence and are not retained after the generation request completes.
• Generated slide content is written back to your Google Slides presentation; we do not keep a copy.
• You can revoke our access to your Google account at any time by visiting your Google account permissions page.

7. Google API Services User Data Policy

• The use of raw or derived user data received from Workspace APIs adheres to the Google User Data Policy, including the Limited Use requirements.
• We explicitly affirm that Google Workspace APIs are not used to develop, improve, or train generalized or non-personalized AI/ML models.
• The Application's use and transfer of information received from Google APIs to any other app adheres to the Google API Services User Data Policy, including the Limited Use requirements.
• Log data: in case of an error while using our app, we collect basic diagnostic data (HTTP status codes, error messages, anonymized request shape) through Cloudflare Workers observability for the purpose of investigating bugs and abuse. This may include IP address, user agent, timestamps, and the configuration of the request — never the body content of your slides or uploaded documents.

8. Personal Information

Our add-on operates inside Google Workspace and therefore requires a Google account for authentication. The default Google authentication flow may expose your email address to our backend solely for the purposes of usage attribution and customer support. Authentication can be withdrawn at any time. Your email address may be used to send communications such as confirmation emails and customer-service updates that are considered necessary to provide the service to you.

9. Business Information

Our add-on logs operational data through Cloudflare Workers observability and Google Cloud Logging. The main purposes of this logging are (a) to improve the add-on and (b) to investigate bugs or abuse. The logged data mainly consists of function-call metadata (e.g. "sidebar opened", "outline generated", "image lookup succeeded"). Additional data (user-input content) is logged only in case of errors and is used solely for debugging — never for any commercial purpose.

10. Google Authentication Scopes

The add-on uses the following OAuth scopes:

11. Contact Us

If you have any questions about this Privacy Policy, please contact us at:

[email protected]