202 points76 comments
Anthropic says commercial Claude chats won't train its models
What is Is my data used for model training??
Anthropic’s privacy page clarifies that inputs and outputs from Claude for Work, the API, and other commercial services are not used to train future models by default. Only feedback submitted via the thumbs-up/down button, or explicit opt-in, may be retained and used for research or model training.
Who it is for
Enterprises and developers using Claude can rely on this policy to keep proprietary data out of training sets, but should still avoid sharing sensitive content unless they opt in.
How to try it
Open https://privacy.claude.com/en/articles/7996868-is-my-data-used-for-model-training