Acceptable Use Policy

1. Scope

This Policy covers:

• Prompts and other instructions you submit.
• Photos you upload.
• Images the Service generates or that you save.

2. Prohibited content

The Service prohibits generating these six categories. Each one is listed separately:

1. Pornography / NSFW. Sexual, nude, or sexually suggestive content, including sexualized images of real people.
2. Violence / gore. Graphic violence, gore, or content that promotes physical harm.
3. Hate speech. Content that attacks or incites hatred against a person or group because of race, ethnicity, nationality, religion, gender, sexual orientation, disability, or a similar trait.
4. Child-unsafe content (CSAM). Any sexual content involving a minor, and any content that sexualizes a child.
5. Deepfakes / impersonation of real people. Face swaps, nudification, intimate-scene compositing, and fabricated statements or actions, whether or not the person consents.
6. Copyright or trademark infringement. Content that copies a protected character, brand, or artwork, or that is presented as official Studio Ghibli work.

2.1 Zero tolerance

The following is removed immediately, the account may be closed, and we may report it to the authorities:

• Child sexual abuse material, or any sexual content involving a minor.
• Content that promotes or helps terrorism, violent extremism, or mass violence.
• Technical instructions for weapons of mass destruction.
• Content that incites genocide or ethnic hatred.
• Content that encourages suicide or self-harm.
• Deepfakes of real people, including face swaps, nudification, intimate-scene compositing, and fabricated statements or actions, whether or not those people consent.

2.2 No exceptions for style

Pornography, graphic violence, and hate speech stay prohibited when the request is framed as art, fiction, anime, or education. Do not present AI output as medical, legal, or financial advice from a licensed professional.

2.3 Other prohibited conduct

• Generating or spreading misinformation, including fabricated events presented as real.
• Impersonating a real person, brand, or organization.
• Infringing intellectual property, publicity, or privacy rights.
• Passing AI-generated content off as a photograph or as human-made work.
• Bulk-generating spam or deceptive commercial content.
• Jailbreak prompts meant to bypass safety filters.
• Commercial use of Output. Personal, non-commercial use is the limit in the Terms of Service.

2.4 Photos of real people

Upload only your own photo, or a photo you have permission to use from the people in it. Uploading someone else’s photo without permission is prohibited. Permission does not allow the deepfake content banned in Section 2.1. The Service does not offer face swap, nudify, or intimate-scene tools.

3. How we moderate

Moderation has three layers: automated screening, human review, and a monthly sample check.

3.1 Automated screening

• Before generation, prompts are screened for prohibited terms, meaning, and intent.
• After generation, the image is screened before it is shown.
• Unusual usage patterns can be blocked.
• Safety systems: OpenAI’s safety filters, plus our own blocklist.

The blocklist covers pornography and NSFW terms, child-unsafe terms, violence, hate speech, self-harm, deepfakes, celebrity and political impersonation, and copyrighted characters, brands, and trademarks. We update it when new abuse shows up. A blocked prompt is not sent for generation. The user sees a notice that the request does not meet this Policy.

3.2 Human review

Content the filters flag, and every user report, is reviewed by UserFirst. Review hours are Monday to Friday, 9:00 a.m. to 5:00 p.m. US Eastern Time. Reports about child safety are handled first.

3.3 Monthly check

At least once a month we sample generated and saved images, check whether the filters missed a prohibited category, and update the blocklist. We do not commission a separate outside audit.

3.4 Severity

4. Reporting

4.1 What to report

• A prompt or image that breaks this Policy.
• Anything involving child safety or terrorism.
• A deepfake or impersonation of you, or a use of your copyright or trademark.

4.2 How reports are handled

• We keep the reporter’s identity confidential.
• Each report is reviewed against this Policy.
• False or abusive reports can themselves be enforced.

4.3 Where to report

Email [email protected] with the subject “Content report”. Include the prompt or image, the account email if you have it, and which category it falls under. This inbox is monitored. There is no separate phone line or in-product report button.

4.4 Response times

The path is: we confirm receipt, classify the report, review it, decide, enforce the decision, and email the reporter. If a case needs longer, we email the reason.

5. Enforcement

6. Appeals

Email [email protected] with the subject “Content appeal” within 14 days of the notice. Include the account email and why the decision should change. We reply within 10 business days and explain the outcome in writing.

7. Changes

For a material change, we email the address on your account at least 14 days before it takes effect and update the date at the top of this page. Continuing to use the Service after that date means you accept the update.