OpenAI publishes privacy policy detailing data collection practices

OpenAI has published a privacy policy that outlines the categories of personal data it collects and how it uses that information.

The company collects account information, user content including prompts and uploaded files, communication information and other information users provide when they interact with events or surveys.

OpenAI also collects technical information from use of its services, including log data, usage data, device information and location data.

The company receives information from partners and public sources used to develop the models that power its services.

OpenAI says it uses personal data to provide, analyse and maintain services, to improve and develop services and conduct research, to communicate with users, to prevent fraud and protect security, and to comply with legal obligations.

Personal data may be disclosed to vendors, affiliates, business-account administrators, counterparties in transactions and government authorities, the policy says.

Data are retained only as long as needed for service provision or legal purposes, the policy states.

ChatGPT temporary chats are kept up to 30 days for safety purposes.

Users may exercise rights including access, deletion, correction, portability, restriction and objection through privacy.openai.com.

Related reading

• OpenAI's new coding platform "helped create itself"
• OpenAI details Codex App Server for agent-based developer tools
• OpenAI’s chip shopping spree is noise, not a Nvidia nightmare

Authorised agent requests can be submitted to dsar@openai.com.

The policy provides instructions for opting out of use of content to train models and for requesting corrections or removals.

The Recap

• OpenAI published a policy describing collected personal data categories.
• Temporary ChatGPT chats are retained for up to 30 days.
• Users can submit privacy requests through privacy.openai.com or dsar@openai.com.