Other
Be Careful What You Tell Your AI Chatbot
Jennifer King and Stanford Institute for Human-Centered Artificial Intelligence, Stanford Institute for Human-Centered Artificial Intelligence (HAI), 2025
Read the originalRead on October 17, 2026
- What it is
- A Stanford HAI news article on a study of the privacy policies of six US frontier AI developers (Amazon, Anthropic, Google, Meta, Microsoft and OpenAI). It reports that all six use users' chat data to train their models by default, and that some keep that data indefinitely.
- How much weight it carries
- Written by the institute that ran the study, and it reports the authors' findings in their own words. It summarizes a peer-reviewed paper in the AAAI/ACM Conference on AI, Ethics, and Society (2025), which was not reread for this guide. The study reads written policies, not how the products behave in practice.
Facts taken from this source
- Stanford HAI reports that Anthropic changed its consumer terms so that conversations with its Claude chatbot are used for training by default, unless the user opts out.
“Last month, Anthropic made a quiet change to its terms of service for customers: Conversations you have with its AI chatbot, Claude, will be used for training its large language model by default, unless you opt out.” (Opening paragraphs)
- In a 2025 Stanford study of the privacy policies of six US AI developers, all six use users' chat data to train their models by default.
“The scholars found all six companies employ users’ chat data by default to train their models” (Section: Blurred Boundaries)
- In the same Stanford study, only some of the six AI developers say they remove personal information before using data for training.
“Some, but not all, of the companies state that they de-identify personal information before using it for training purposes.” (Section: Blurred Boundaries)