Our Mission & Approach to Safety
Powerful AI requires thoughtful development. As an open-source project, we are committed to building safe and beneficial text-to-video technology through community collaboration.
Safeguards for the Hosted Service
OpenSora 2 combines automated checks, provider safeguards, account controls, and human review of reports. These controls apply to free and paid workflows, including private generations.
- Pre-generation checks: Prompts and supported reference images are checked before generation, and configured moderation fails closed when a required check is unavailable.
- Prohibited content: Pornography, sexual exploitation, content involving minors, non-consensual intimate imagery, harmful deepfakes, impersonation, hateful or violent abuse, and illegal content are not permitted.
- Consent and rights: Only upload people, media, voices, brands, and other assets you have permission to use.
- Reports and appeals: Report a content URL or task ID through support tickets or email support@opensora2.com. Do not attach illegal material. We can restrict or remove content and accounts, and appeals are reviewed through the same ticket.
