Companies Reserving AI-Training Rights in Their Terms — TOSTracker Report

8 min read Original article ↗

Finding

As of July 2, 2026, TOSTracker's archive contains verified AI-training-rights clauses — provisions in which a company reserves the right to use user-provided content or data to train artificial-intelligence or machine-learning models — in the terms of service or privacy policies of 23 companies, listed below with the verbatim clause text, the source document, and the date each was captured.

23

Companies with a verified AI-training clause

55,156

Legal documents tracked

206,292

Version snapshots archived

Scope & definitions

A clause is counted only where the company reserves the right to use user content or data to train its own (or a third party's) AI/ML models. Excluded: statements that a company does not train on user data; prohibitions on users training AI on the company's content; and generic "we use AI features" descriptions.

Tags: opt-out / opt-in — the terms document a choice mechanism; de-identified — the clause limits use to de-identified or aggregated data; shared with third-party AI partners — content is provided to outside AI developers; public/collected data — training draws on publicly available or generally collected data; free/trial tiers — the clause is scoped to specified plans.

Verified clauses (23 companies)

⬇ Download the dataset (CSV, with archive URLs + SHA-256 hashes)

CompanyDocumentVerbatim clause (excerpt)Live sourcePermanent archive
Anthropic Terms of Service “We may use Materials to provide, maintain, and improve the Services, including training our models, unless you opt out of training through your account settings.”

opt-out

live source
cap. 2026-04-03
Archived snapshot →

SHA-256 9d7d3426ae9b4107…

Brevo Terms of Service “Brevo may use Content and Usage Data to enrich and train its AI Features for the sole purpose of improving the performance of the Services provided to You…” live source
cap. 2026-05-01
Archived snapshot →

SHA-256 d93240a0e6b54dbc…

Domino's Pizza Privacy Policy “We collect and use information to develop new features, enhance functionality, and train our models and algorithms.” live source
cap. 2026-04-08
Archived snapshot →

SHA-256 02551907a9fbe382…

GitHub (Microsoft) Terms of Service “If you provide your private repository content as Input to AI Features, we may use that Input to provide, develop, train, and improve the Service, including AI Features.” live source
cap. 2026-06-19
Archived snapshot →

SHA-256 c36bc372f7fb63b4…

Google / YouTube Privacy Policy “…we may collect information that's publicly available online or from other public sources to help train Google's AI models and build products and features like Google Translate, Gemini Apps, and Cloud AI capabilities.”

public/collected data

live source
cap. 2026-04-10
Archived snapshot →

SHA-256 b648850449f2a374…

HubSpot Privacy Policy “…we may process personal data to develop, support, and improve HubSpot AI features and to train our AI models and similar products and services that rely on machine learning.” live source
cap. 2026-04-15
Archived snapshot →

SHA-256 83ae9db004a4f2b7…

Indeed Terms of Service “…we use such data about your particular activity, communication, and materials to develop, train, build, and use statistical models, including artificial intelligence and machine learning models.” live source
cap. 2026-04-04
Archived snapshot →

SHA-256 9c38f694f450f3fb…

LinkedIn Privacy Policy “We may use your personal data to improve, develop, and provide products and Services, develop and train artificial intelligence (AI) models…” live source
cap. 2026-04-08
Archived snapshot →

SHA-256 01167cb8c31e906a…

OpenAI Terms of Service “If you do not want us to use your Content to train our models, you have the option to opt out by updating your account settings.”

opt-out

live source
cap. 2026-04-05
Archived snapshot →

SHA-256 4cea697e0ec46e5f…

Oscar Health Terms of Service “Oscar may utilize any User Submission (in a de-identified and anonymized form) to train, optimize, ground or otherwise enhance its AI Technology.”

de-identified

live source
cap. 2026-02-05
Archived snapshot →

SHA-256 757fdb85f5cf8256…

PayPal Privacy Policy “We may use Personal Information to train our artificial intelligence (AI) models that power our Services and help us deliver more secure, efficient, and personalized services.” live source
cap. 2026-01-31
Archived snapshot →

SHA-256 c2b5c49a4e02a955…

Paycor Terms of Service “Paycor may use information you provide to develop, train, and/or improve our services, business processes, AI…” live source
cap. 2026-04-03
Archived snapshot →

SHA-256 e04581c5deca88e0…

Prezi Terms of Service “We may share your reusable Public User Content with our AI third-party partners who may use this content for their own commercial purposes, including without limitation, to train and improve their AI algorithms and models…”

shared with third-party AI partners

live source
cap. 2026-03-17
Archived snapshot →

SHA-256 4a5675c358f27d79…

Reddit Terms of Service “…this license includes the right to use Your Content to train AI and machine learning models, as further described in our Public Content Policy.” live source
cap. 2026-05-01
Archived snapshot →

SHA-256 cdb5ac788c838a49…

Rumble Terms of Service “…by submitting Content to the Rumble Service, you grant to Rumble the right to use the Content to train AI and machine learning models…” live source
cap. 2026-02-11
Archived snapshot →

SHA-256 380d37aa8959d897…

RunwayML Terms of Service “You acknowledge that Inputs and Outputs may be used by the Company to train and improve its AI models, algorithms and related technology, products and services…” live source
cap. 2026-03-17
Archived snapshot →

SHA-256 1f785e6d7589ad1c…

Together AI Terms of Service “…you have the ability to control how your data is handled by choosing "No" when asked if you want to store prompts or allow your data to train models.”

opt-out

live source
cap. 2026-06-19
Archived snapshot →

SHA-256 03bf7ee33c6efd33…

Upwork Terms of Service “Opted-in users grant Upwork a limited license to use User Content including Work Product that they exchange through the platform to train artificial intelligence (AI) tools…”

opt-in

live source
cap. 2026-03-16
Archived snapshot →

SHA-256 90f154f0bb7c4fe8…

VSCO Terms of Service “This license includes the right to use certain Content… to develop, train, and improve AI or machine learning models as further described in our Creator Content Standards.” live source
cap. 2026-02-25
Archived snapshot →

SHA-256 31ad7878da4f5f9f…

Vercel Terms of Service “…if you are on a Hobby plan or trial Pro plan, you agree that we may use Your Content to train our artificial intelligence ("AI") and machine learning models…”

free/trial tiers

live source
cap. 2026-03-31
Archived snapshot →

SHA-256 09f76a489f095fab…

Virta Health Terms of Service “We may use de-identified and aggregated data derived from your use of the Service to train, develop, and improve our own internal AI models. This data is de-identified in accordance with HIPAA standards…”

de-identified

live source
cap. 2026-04-15
Archived snapshot →

SHA-256 448cc04198a78751…

X (Twitter) Privacy Policy “We may use the information we collect and publicly available information to help train our machine learning or artificial intelligence models for the purposes outlined in this policy.”

public/collected data

live source
cap. 2026-02-03
Archived snapshot →

SHA-256 58d9fc760c20c4ee…

Zapier Terms of Service “Zapier may … use such Derived Data to operate, enhance, improve, and develop Zapier products or services, including through model training. You may opt out of providing Zapier with such permission for Derived Data…”

opt-outde-identified

live source
cap. 2026-04-12
Archived snapshot →

SHA-256 a91acf7c61548d2c…

Each “Archived snapshot” is a permanent, immutable capture with its own citeable URL and SHA-256 content hash — verifiable independently. “Scorecard” grades the document’s user-friendliness; “evidence exhibit” is a court-ready, printable single-page record of the exact version.

How this list was produced

TOSTracker archives Terms of Service, privacy policies, and related documents for 55,156 organizations (44,505 active), retaining 206,292 SHA-256-hashed version snapshots. A rule-based extractor first flagged 125 documents as containing AI-training language. Each flagged document was reviewed against its full source text: 42 documents across 23 companies were confirmed as genuine AI-training-rights clauses, and 83 were removed as false positives — chiefly denials ("we do not train on your data"), prohibitions barring users from training AI on the company's content, and incidental mentions of AI features. Only confirmed clauses appear above.

Clause text is quoted from TOSTracker's capture on the listed date. Ellipses (…) mark omitted surrounding text; no words within a quotation are altered. Policies change — each row links to the live source for reconfirmation. Provisions are stated as reserved rights ("may"): they record what a company's terms permit, not a determination of what it is doing at any given moment.

Citation

Leahey, A. Companies Reserving AI-Training Rights in Their Terms. TOSTracker Case Studies, July 2, 2026. https://tostracker.app/analysis/ai-training-rights

This report is generated from publicly available terms of service, privacy policies, and other legal documents collected by TOSTracker. It is provided for informational and research purposes only and does not constitute legal advice. TOSTracker is an independent monitoring service.