Anthropic Consumer Terms and Privacy Policy Updates

Anthropic has updated its Consumer Terms and Privacy Policy to introduce an opt-in mechanism for using consumer data to train future Claude models and enhance safety safeguards. This change allows users on Free, Pro, and Max plans to choose whether their interactions are used to improve model reasoning, coding, and analysis skills, as well as to refine the detection of harmful content.

Data Usage and Model Improvement

Anthropic is implementing a choice for users to allow their data to be used for the purpose of improving Claude's capabilities and strengthening safeguards against scams and abuse. According to the announcement, participating users help the company improve model safety by making systems for detecting harmful content more accurate and reducing the false-positive rate for harmless conversations.

Beyond safety, this data usage is intended to improve future Claude models in specific technical domains, including:

• Coding
• Analysis
• Reasoning

Scope of Application

These policy updates apply specifically to consumer-facing plans and tools. The updates affect users on the following plans:

• Claude Free
• Claude Pro
• Claude Max
• Claude Code (when used via accounts associated with the above plans)

These updates do not apply to services governed by Commercial Terms, including:

• Claude for Work
• Claude for Government
• Claude for Education
• API usage (including access via third parties such as Amazon Bedrock and Google Cloud’s Vertex AI)

Data Retention and Privacy Controls

For users who opt-in to model training, Anthropic is extending the data retention period to five years to support model development and safety improvements. This five-year retention period also applies to feedback submitted regarding Claude’s responses.

Users who do not opt-in to model training will remain under the existing 30-day data retention period. To protect privacy, Anthropic utilizes automated processes to filter or obfuscate sensitive data and states that it does not sell user data to third parties.

If a user deletes a conversation with Claude, that conversation will not be used for future model training.

Implementation Timeline and User Action

Existing users must accept the updated Consumer Terms and make their training preference selection by October 8, 2025. After this date, a selection on the model training setting is required to continue using Claude.

New users can select their preference during the signup process, while existing users are presented with a choice via an in-app notification. Preferences can be managed and changed at any time within the Privacy Settings.