Privacy

This page is written to describe how the site actually behaves, but it is not legal advice and it has not been reviewed by a lawyer. BuildTuber is operated by Sahadya LLC, 1309 Coffeen Avenue, Sheridan, WY 82801. Privacy questions and deletion requests can go to ethan@buildtuber.com.

The site is operated from Wyoming, USA. The California Consumer Privacy Act does not apply to it, because that law reaches only businesses above revenue and data-volume thresholds this one is nowhere near. The EU and UK GDPR do apply to subscribers living in those places, whatever the site is run from, so this page is written to meet the stricter of the two rather than the more convenient one.

The short version

This site sells nothing directly, sets no advertising cookies, and runs no third-party tracking scripts. The only personal data it asks for is an email address, and only if you choose to hand one over.

The BuildTuber add-on for Google Sheets

BuildTuber's video-pipeline product is a Google Workspace Marketplace add-on, written in Google Apps Script, that runs inside your own Google account against your own spreadsheet. There is no BuildTuber-operated server in the pipeline. The four sections below state what Google user data the add-on accesses, how that data is used, who it is shared with, and how it is protected, retained and deleted.

What Google user data the add-on accesses

The add-on requests these scopes, and accesses only the Google user data each one covers:

• Spreadsheets (.../auth/spreadsheets). Read and write access to the single spreadsheet the add-on is running in: your queue of videos, the narration lines you paste into the storyboard sheet, your configuration values (including the third-party API keys described below), and a hidden sheet used to checkpoint progress between runs. The add-on's engine also runs on a time-based trigger, which is why the broader spreadsheets scope is requested rather than a current-document-only scope.
• Drive files (.../auth/drive.file). Access limited to files and folders the add-on itself creates. It creates one output folder per video containing the reference images you upload, generated frames, voiceover audio, and a render manifest. Reference images are uploaded from your own computer through the add-on's sidebar and written to Drive as new files; the add-on never reaches into files that were already there.
• Add-on interface (script.container.ui), external requests (script.external_request), triggers (script.scriptapp), and script storage (script.storage). These four scopes grant no access to your Google user data. They let the add-on draw its sidebar and menu, call the third-party providers listed below, schedule its own background runs, and keep its own progress state and licence status in Apps Script's private key-value store.

The add-on does not access your Gmail, Calendar, Contacts, photos, profile information, or any other Google service. It does not request your name or email address, and does not read your Google account identity. It never enumerates, lists, opens, reads, or modifies any other spreadsheet, file, or folder that already existed in your Drive.

How that Google user data is used

Google user data is used solely to provide the feature you asked for: turning narration you wrote into the image, audio and timing assets for one video. Spreadsheet contents are read to know what to generate and written back to record what was generated. Drive is used to store those generated assets and the reference images you upload. Nothing is used for advertising, profiling, or any purpose unrelated to running the pipeline in your own spreadsheet.

BuildTuber does not use Google user data to train, retrain, fine-tune, or otherwise improve any artificial intelligence or machine learning model. BuildTuber operates no model of its own and holds no copy of your data to train on. Where the add-on sends your content to a third-party AI provider on your behalf, it does so under your own account with that provider, using your own API key, as described in the next section.

Who the data is shared with

BuildTuber itself receives no Google user data. It is not shared with, transferred to, or stored by BuildTuber or Sahadya LLC, and it is never sold, rented, or transferred to any advertiser, data broker, or other third party.

BuildTuber is bring-your-own-key. You paste your own Google Gemini, Runware, and ElevenLabs API keys into your spreadsheet's configuration, and the add-on calls those providers directly from your own Apps Script execution, under your own accounts:

• Google Gemini receives your storyboard lines and reference images, to write image prompts and a visual style profile.
• Runware receives those image prompts, and the reference images uploaded directly as image data, to generate frames. Generated frames are returned inline as image data rather than as a hosted link, so Runware does not retain an output URL for them, and Runware is never given a link to anything in your Drive.
• ElevenLabs receives your narration text, to generate voiceover audio.

Each provider processes what you send it under its own terms and privacy policy and under your own contract with them. Your API keys, your storyboard, your narration, and the images and audio these calls return never pass through any BuildTuber-operated system.

Third-party AI providers and model training

Because the add-on sends content from your spreadsheet to these providers, it matters whether they train on it. Two of them train by default, so BuildTuber requires the following configuration. These are setup requirements for the product, not suggestions.

• Google Gemini - a billing-enabled (paid) API key is required. Google's Gemini API terms state that on the unpaid tier it uses submitted content "to provide, improve, and develop Google products and services and machine learning technologies." On the paid tier Google states it "doesn't use your prompts ... or responses to improve our products." The paid tier is pay-as-you-go and billed per token; it is not a subscription and carries no monthly fee. A free-tier key must not be used with this add-on.
• ElevenLabs - model training must be turned off on your account. On the Free, Creator, Pro and Scale plans, ElevenLabs uses submitted content for model training unless you opt out under Settings > Data use. The opt-out applies only to content submitted after you set it, so it should be set before you generate anything. Enterprise accounts have training off by default and can additionally enable zero-retention mode in the add-on's configuration.
• Runware - does not train on what you send. Runware's terms commit that it will not use Training Data or Trained Models to train, fine-tune or develop its own or any third party's models. Because that clause is written around data uploaded for model customisation, and BuildTuber uses ordinary image inference rather than customisation, BuildTuber asked Runware directly. Runware confirmed the commitment also covers inference: the prompts and reference images sent for image generation are not used for training.

Models that are not run by Runware

Runware serves some models on its own infrastructure and routes others to the company that made the model. Where a model is routed onward, that company's terms apply to the prompt and reference images for that request. The models BuildTuber offers, and where each one runs:

• FLUX.2 Klein runs on Runware's own infrastructure, under the Runware commitment above. It is the add-on's default model.
• Grok Imagine Image is xAI's. xAI's API documentation states that "xAI never trains on your API inputs or outputs without your explicit permission."
• Qwen-Image is Alibaba's. Alibaba Cloud Model Studio's privacy notice states that Alibaba Cloud "will never use your data for model training."
• Seedream is ByteDance's. ByteDance's platform does not train on customer data unless the account holder has separately signed its data authorization agreement. BuildTuber does not hold that account and cannot verify its status, so if you want a documented no-training position on every request, use one of the models above.

Model choice is yours and is set in the add-on's configuration sheet. The "Optimal" setting picks between FLUX.2 Klein and Grok Imagine per frame, so it stays within the two entries above. No Google user data reaches any of these providers except the storyboard text and reference images you put in your own spreadsheet and ask the add-on to render.

The use of raw or derived user data received from Google Workspace APIs adheres to the Google User Data Policy, including the Limited Use requirements, and such data is not used to create, train, or improve any generalized or foundational AI/ML model.

The one BuildTuber-operated component the add-on contacts is a small Cloudflare Worker that validates your licence. That request carries only your licence key and a product identifier. It carries no Google user data: no spreadsheet content, no Drive file, no Google account identity. Nothing else in the add-on communicates with a BuildTuber server.

BuildTuber'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.

How the data is protected, retained, and deleted

Google user data stays in your Google account. It is held in your spreadsheet and your Drive, under Google's own encryption at rest and in transit and under your own account's security settings, including whatever two-factor authentication and sharing controls you have in place. Because BuildTuber keeps no copy, there is no BuildTuber database that can be breached, subpoenaed, or leaked.

All outbound calls the add-on makes - to Gemini, Runware, ElevenLabs, and the licence Worker - are made over HTTPS. Your third-party API keys are stored as ordinary cells in your own spreadsheet's configuration sheet, which means they are protected by that spreadsheet's sharing settings and are visible to anyone you give access to it. Keep the sheet private, and treat sharing it as sharing those keys. They are never transmitted to BuildTuber.

Files the add-on generates in your Drive are set to "anyone with the link can view," so that the Colab notebook that renders your video can download them, and so that generated frames can preview inside the spreadsheet. The links are unguessable and unlisted, and BuildTuber does not index, publish, or share them. You can tighten or remove those permissions yourself in Drive at any time once a video is rendered.

Retention and deletion. BuildTuber retains no Google user data, so there is no retention period on our side and no separate copy to request the deletion of. Your data persists only for as long as you keep the spreadsheet and output folders in your own Drive. Deleting them, and emptying your Drive trash, removes the data completely. You can revoke the add-on's access at any time from your Google Account permissions page, or uninstall it from the Google Workspace Marketplace, either of which stops all access immediately.

Email, if you subscribe

The subscribe form is a plain HTML form that posts to Kit (formerly ConvertKit), which handles the list. Kit receives your email address and, as part of delivering the messages, technical details such as the time you subscribed and whether you opened an email. It is used for one thing: sending the build guides and occasional notes about what is being published.

Every email carries an unsubscribe link. Using it removes you from the list. You can also ask for your record to be deleted outright at the contact address below.

Analytics

Page views are counted with Umami, which is cookieless. It records the page you landed on, the referring site, and coarse details like country and device type. It does not set cookies, does not use browser fingerprinting, and does not build a profile that follows you between sites. Nothing it collects identifies you, which is why this site shows no cookie banner.

Search engine verification tags from Google Search Console and Bing Webmaster Tools are present in the page source. These are inert meta tags that prove ownership of the domain. They do not track anything.

Purchases

Paid products are sold and delivered through Whop. When you buy something you leave this site and transact with Whop directly. This site never sees, handles, or stores your payment details, and no card information passes through it at any point. Whop is the data controller for anything you enter during checkout, under its own privacy policy.

Hosting

The site is static and hosted on Netlify. Like any web host, Netlify processes standard server request data, including IP addresses, in order to serve pages and absorb abuse.

What is never done

• No advertising or retargeting pixels.
• No selling or renting of email addresses.
• No embedded social widgets that load before you click them.
• No cross-site tracking of any kind.

Lawful basis under EU and UK GDPR

For subscribers in the EU or the UK, the lawful basis for holding an email address is consent. It is given by submitting the form and then clicking the confirmation link that arrives afterwards, which is why that second step exists. It is withdrawn with the unsubscribe link at the foot of every email, and withdrawing it does not undo anything sent while it stood. No other lawful basis is relied on, because nothing else here collects personal data.

Your rights

Depending on where you live you may have the right to ask for a copy of the data held about you, to have it corrected, or to have it deleted. For this site that will almost always mean one row in a mailing list. Ask and it gets handled.

Contact

Privacy questions and deletion requests go to ethan@buildtuber.com. BuildTuber is operated by Sahadya LLC, 1309 Coffeen Avenue, Sheridan, WY 82801.