Anthropic · Anthropic Privacy Policy

AI Model Training Using Inputs and Outputs

High severity
Share 𝕏 Share in Share 🔒 PDF
Watch Anthropic Get alerts when this provision or policy changes.
Watch — $9.99/mo

Why it matters (compliance & risk perspective)

Your private conversations with Claude — including sensitive personal topics — may become training data for AI models, and the opt-out has significant exceptions that most users will not anticipate.

Consumer impact (what this means for users)

Anthropic collects your conversation content (prompts and outputs), device identifiers, usage patterns, and payment information when you use Claude.ai, and by default uses your conversations to train its AI models. Users who opt out of model training may still have their conversations used if those conversations are flagged for safety review, which is a significant limitation on the opt-out right. You can opt out of having your conversations used for model training by navigating to your account settings on Claude.ai, and you can delete individual conversations which are removed from your history immediately and purged from Anthropic's back-end within 30 days.

How other platforms handle this

Dropbox Medium

When transferring data from the European Union, the European Economic Area, the United Kingdom, and Switzerland, Dropbox relies upon a variety of legal mechanisms, such as contracts with our customers and affiliates, Standard Contractual Clauses, the EU-U.S. Data Privacy Framework, the UK Extension ...

Skillshare Medium

In connection with any reorganization, restructuring, merger or sale, or other transfer of assets, we will transfer information, including personal information, provided that the receiving party agrees to respect your personal information in a manner that is consistent with our Privacy Policy.

Stripe Medium

Depending on the context, 'you' might be an End Customer, End User, Representative, or Visitor. End Customers interact with Stripe's services through Business Users (e.g., when purchasing from a merchant). For End Customers, the Business User is the primary data controller and Stripe acts as a data ...

See all platforms with this clause type →

This clause could change without notice.

Get alerted when Anthropic updates this policy — with plain-language summaries and severity ratings.

Watch Anthropic Need compliance memos? Professional →
View original clause language
We may use your Inputs and Outputs to train our models and improve our Services, unless you opt out through your account settings. Even if you opt-out, we will use Inputs and Outputs for model improvement when: (1) your conversations are flagged for safety review to improve our ability to detect harmful content, enforce our policies, or advance AI safety research, or (2) you've explicitly reported the materials to us (for example via our feedback mechanisms).

Applicable regulations

BIPA
Illinois, USA
CCPA/CPRA
California, USA
COPPA
United States Federal
CAN-SPAM
United States Federal
DMA
European Union
FCRA
United States Federal
GDPR
European Union
GLBA
United States Federal
HIPAA
United States Federal
UK GDPR
United Kingdom

Provision details

Document information
Document
Anthropic Privacy Policy
Entity
Anthropic
Document last updated
April 29, 2026
Tracking information
First tracked
March 6, 2026
Last verified
April 28, 2026
Record ID
CA-P-002561
Document ID
CA-D-00012
Evidence Provenance
Source URL
Wayback Machine
SHA-256
55f589f5c2a5a187a9d045dc6c7e4954a2dbf9ac00fb6e3ea782dbcf9ad69387
Verified
✓ Snapshot stored   ✓ Change verified
How to Cite
ConductAtlas Policy Archive
Entity: Anthropic | Document: Anthropic Privacy Policy | Record: CA-P-002561
Captured: 2026-03-06 20:00:36 UTC | SHA-256: 55f589f5c2a5a187…
URL: https://conductatlas.com/platform/anthropic/anthropic-privacy-policy/ai-model-training-using-inputs-and-outputs/
Accessed: May 4, 2026
Classification
Severity
High
Categories

Other risks in this policy

Related Analysis

Don't miss changes to this clause.

Anthropic has updated this policy before. Get alerted on the next change.

Watch Anthropic

Frequently Asked Questions

What does Anthropic's AI Model Training Using Inputs and Outputs clause do?

Your private conversations with Claude — including sensitive personal topics — may become training data for AI models, and the opt-out has significant exceptions that most users will not anticipate.

Is ConductAtlas affiliated with Anthropic?

No. ConductAtlas is an independent monitoring service. We are not affiliated with, endorsed by, or sponsored by Anthropic.