AIPolicyLog

Head-to-head · checked 2026-11-18

ChatGPT vs Claude: which trains on your data?

Both train on individual consumer accounts by default with an opt-out, and neither trains on business or API data by default. Both say deleted chats leave their systems within 30 days. The documented difference is that Anthropic may keep de-identified data up to 5 years in its training pipelines when you allow model improvement.

Both columns come from each vendor’s own pages, checked 2026-11-18 (America/Toronto (UTC−5)). See the full quotes on the OpenAI / ChatGPT and Anthropic / Claude pages.

ChatGPTClaude
Short answerYes, by default on ChatGPT Free, Plus and Pro: OpenAI says data sharing is enabled by default on those personal workspaces, and you can turn it off; ChatGPT Business, Enterprise, Edu and the API are not used for training by default.Yes, on consumer Claude plans (Free, Pro, Max and Ultra) unless you opt out: Anthropic’s privacy policy says it may use your inputs and outputs to train its models; Claude for Work, the API and Claude Gov are not used for training by default.
Trains by default?Yes (Free, Plus, Pro)Yes, unless you opt out (consumer)
Personal accountsOn personal Free, Plus and Pro workspaces, eligible content is used to improve the models unless you turn off “Improve the model for everyone” or choose “Do not train on my content” in OpenAI’s Privacy Portal.Consumer inputs and outputs may be used for training unless you turn off Model Improvement in your Privacy Settings. Anthropic says turning it off means it will not use your previous or new chats or coding sessions for future model training. Even after opting out, Anthropic says conversations flagged for safety review, and material you explicitly report (for example through feedback), can still be used.
Opt-outYes, in settings · stepsYes, Model Improvement setting · steps
Business, enterprise and APIChatGPT Business, Enterprise, Edu, Healthcare and Teachers, and the API Platform, are not used for training by default. The API has an opt-in for sharing data with OpenAI, and organizations with Zero Data Retention cannot opt in.Commercial products (Team, Enterprise, API, third-party platforms and Claude Gov) are not used for training by default. An organization can opt in, for example through the Development Partner Program.
RetentionDeleted chats are removed from OpenAI’s systems within 30 days (with security and legal exceptions); temporary chats may be kept up to 30 days. API inputs and outputs may be kept up to 30 days, except for endpoints with Zero Data Retention.Deleted consumer chats leave Anthropic’s back-end systems within 30 days. If you allow model improvement, de-identified data may be kept up to 5 years in training pipelines. Chats flagged for Usage Policy violations are kept up to 2 years. Commercial chats follow the organization’s retention setting.
Zero data retentionAPI only, for eligible customersAPI, by arrangement

Sources

More comparisons

Get an email when ChatGPT or Claude changes its data policy

Free email alerts when AI tools change their data, training or retention terms. No spam. Unsubscribe anytime.