使用条款

Content safety policy

Content policy effective October 6, 2026.

This policy applies to uploaded images and videos, custom text, generated results, and use of the service. Do not submit, generate, or distribute:

• Violence and gore: graphic injury, mutilation, torture, dismemberment, or content encouraging or glorifying real-world violence.
• Hate speech: threats, dehumanization, incitement to violence, or discrimination targeting people based on race, ethnicity, nationality, religion, caste, sex, gender identity, sexual orientation, or disability.
• Sexual or exploitative content: pornography and other NSFW content, explicit nudity or sexual acts, sexual content involving minors, sexual violence, nonconsensual intimate imagery, or attempts to undress a person.
• Child sexual abuse material (CSAM): any sexual depiction or exploitation of a minor, including generated or altered material.
• Deepfake impersonation: creating or using synthetic media to impersonate another real person or falsely portray their words or actions.
• Non-consensual intimate imagery (NCII): creating, altering, or sharing intimate imagery of a person without their consent.
• Abuse and rights violations: targeted harassment, deceptive impersonation, unauthorized use of a person's likeness, exploitation, or content violating privacy, copyright, trademarks, or other rights.
• Illegal activity: assistance with terrorism, trafficking, or other unlawful harm; evasion of safety controls is also prohibited.

How we review content

Before a generation starts, we screen submitted text using a local child-safety rule and Waffo's automated prompt screening. A blocked or uncertain result does not proceed. If screening is unavailable, the request stops and may be retried later. These pre-generation stops do not charge credits.

Our media pipeline also checks supported uploaded and generated images and videos for explicit nudity. These automated checks are limited and do not detect every violation. Violence, gore, hate, consent concerns, and other suspected violations can be reported for human review; we do not claim that every result is manually reviewed or that automated nudity checks detect violence or hate.

Enforcement and appeals

Authorized administrators review screening records and reports, investigate the supplied information, and record their decisions and reasons. Depending on the evidence and seriousness, we may refuse a request or suspend generation access. A report alone does not automatically suspend another user. Accounts with a content restriction cannot submit new generation requests; billing, subscription cancellation, and appeals remain available. We may restore access after review of an appeal. This process does not automatically forfeit purchased credits or cancel recurring subscriptions; billing and refunds remain subject to the payment and refund terms.

Report a concern

Use 举报内容 without signing in. Select a category and provide the affected link or request reference, what happened, and an optional contact email. Do not upload or reproduce illegal or graphic material. You receive a receipt number when the report is saved. To appeal, select “Appeal a moderation decision” and include the relevant receipt or account reference. We record reports and review outcomes in a restricted administrator queue.

We provide an initial human response within 3 business days when a contact email is supplied, including for photo or likeness misuse. Reports involving suspected CSAM, NCII, or imminent harm are prioritized. If an investigation needs longer, we explain the next steps and provide an update; this is a response commitment, not a guarantee that every investigation finishes within that period.