Usage Policies

This document is OpenAI's Usage Policy (Acceptable Use Policy), which governs permissible and prohibited uses of OpenAI's models, APIs, products, and services across all platforms including ChatGPT, Sora, Codex, and the developer API. The agreement states that users and operators must not use OpenAI's services for a defined set of prohibited activities including generating content that sexualizes minors, facilitating weapons capable of mass casualties, creating cyberweapons, generating disinformation to undermine elections, and building surveillance tools that violate civil rights. The policy establishes a tiered operator-user framework in which API operators bear responsibility for ensuring downstream user compliance, and the terms authorize OpenAI to take enforcement action including suspension or termination for violations, with an appeals process referenced for affected users. The document engages with regulatory frameworks governing AI systems, child safety, cybersecurity, and content moderation, including the EU AI Act, COPPA, the FTC Act, and emerging national AI governance requirements, though the applicability of specific obligations depends on the jurisdiction and whether a deployment qualifies as a high-risk AI use case. Compliance teams should note that the policy places affirmative due diligence obligations on API operators to restrict prohibited uses by their end users, which may require contract amendments, user consent audits, and content moderation infrastructure reviews.