Acceptable Use Policy

Acceptable Use Policy

Last Updated: October 15, 2026

This Acceptable Use Policy applies to Text Photo AI, including prompts, uploaded reference images, generated outputs, account activity, credits, and paid subscriptions.

Text Photo AI is an independent AI image generation and editing service. Outputs may be AI-generated or AI-assisted. You are responsible for using the service lawfully, respecting third-party rights, and reviewing generated content before publishing or sharing it.

1. Independent Product and AI Disclosure

Text Photo AI is operated independently and is not affiliated with, endorsed by, or sponsored by Seedream, ByteDance, APIMart, Waffo Pancake, or any third-party model provider unless expressly stated.

You must not present Text Photo AI outputs as guaranteed factual evidence, official documents, verified identity records, or human-created content where an AI-generated disclosure is legally or contextually required.

2. Prohibited Content and Uses

You may not use Text Photo AI to create, upload, request, generate, distribute, sell, or promote the following content or activity.

2.1 Zero-Tolerance Content

• CSAM, child sexual exploitation, grooming, sexual content involving minors, or any content that harms children.
• Terrorism, violent extremism, terrorist recruitment, praise, propaganda, operational support, or instructions.
• Weapons of mass destruction, including technical instructions, procurement guidance, or operational assistance.
• Real-person deepfakes, regardless of consent: face swaps or identity replacement; nudification or sexualization; fabricated kissing, hugging, or other intimate interactions; and fabricated speech, actions, scenes, or events in images, video, or audio. Public figures and ordinary people are covered, whether using an existing group photo or separate portraits.
• Content that incites genocide or ethnic hatred.
• Non-consensual intimate imagery or sexualized images of real people.
• Content that promotes suicide, self-harm, serious violence, or credible threats against people or protected groups.
• Doxxing, stalking, impersonation for fraud, or privacy-invasive identity abuse.

2.2 Prohibited Image Content and Other Uses

• NSFW, adult, pornographic, sexually explicit, sexually suggestive, fetish, or escort-related content.
• Deepfake, face-swap, face manipulation, likeness cloning, or identity deception involving real people, regardless of consent.
• Harmful, violent, threatening, harassing, hateful, abusive, discriminatory, or extremist content.
• Graphic gore, torture, cruelty, or content designed to shock, intimidate, or incite harm.
• Fraud, phishing, scams, spam, malware, credential theft, evasion of safety controls, or other cyber abuse.
• Content that infringes copyright, trademark, privacy, publicity, likeness, contractual, or other rights.
• Illegal goods, regulated goods, weapons, drugs, gambling, financial fraud, medical misinformation, or other restricted activities.
• Attempts to bypass moderation, rate limits, account restrictions, payment controls, geographic restrictions, or technical safeguards.

3. Content Moderation

Text Photo AI uses a layered content moderation process for AI image workflows. These controls are designed to reduce abuse, but they do not replace your own legal and policy responsibilities.

3.1 Input Screening

Text Photo AI screens user prompts before image generation using Waffo Prompt Sift. Only prompts that pass moderation are sent to the configured AI image provider.

Only an allow verdict permits generation. A block verdict stops generation; a review verdict pauses it and asks you to retry later. If moderation is unavailable, generation fails closed. These attempts do not consume image-generation credits.

3.2 Uploaded Images and References

You may upload only images that you have the right to use. For photos or voice recordings of real people, you may upload only your own material or material for which every depicted or recorded person has expressly authorized the upload and processing. Uploading another person's photo or voice without their consent is prohibited. Source-material authorization does not permit prohibited deepfakes. Do not upload private, confidential, regulated, explicit, sexually suggestive, harmful, or rights-infringing images.

Uploaded reference images and related account activity may be reviewed when they are reported, flagged, associated with abuse patterns, or needed to enforce this policy.

3.3 Generated Output Review

We review generated images using APIMart’s omni-moderation-latest endpoint after generation and before display or download. We send each output image URL to APIMart for image retrieval and safety review. This applies to every available model on APIMart and Replicate, including image edits and layer outputs. Every image in a task must receive an explicit approval before the task’s outputs are saved to configured public storage or released to users. Rejections, timeouts, service errors and unclear responses fail closed; the task fails and follows the generation-credit refund process. Historical images without a matching approval record remain unavailable through generation results, history and downloads. Prompt screening alone and reviews triggered by later reports do not replace this output review.

We retain the output review status, time, returned request IDs and category flags with the task, subject to our Privacy Policy. Automated safety checks do not establish ownership, permission to use a real person’s likeness, or compliance with every copyright or trademark requirement.

We review reported outputs, flagged activity, billing abuse signals, repeat violations, and other abuse indicators. We may remove, restrict, withhold, or disable access to generated outputs after generation if they violate this policy, applicable law, payment network rules, or Waffo Pancake review requirements.

3.4 Human Review and Safety Logs

When content is reported or enforcement is needed, our support and safety review process may evaluate prompts, uploaded images, generated outputs, account records, payment records, moderation results, and relevant technical logs.

We retain moderation and enforcement records only as needed for service operation, legal compliance, dispute handling, safety review, and abuse prevention, as described in our Privacy Policy.

4. Severity Levels

5. Reporting Channels

If you see content, prompts, generated outputs, account activity, or payments that may violate this policy, contact us through the channel below.

• Safety, abuse, content report, billing abuse, or appeal email: support@textphotoai.com

Please include the account email if available, the relevant prompt or output URL, screenshots if appropriate, the reason for the report, and any urgency such as child safety, violence, fraud, or privacy risk.

6. Response and Processing Timelines

Complex cases, missing evidence, external provider review, payment disputes, legal requests, or safety escalations may require additional time. We prioritize child safety, non-consensual sexual content, credible threats, terrorism, and severe fraud.

7. Enforcement Actions

If we believe this policy has been violated, we may take one or more of the following actions:

• Block prompts, reject generation requests, or prevent uploads.
• Remove, restrict, or withhold access to generated outputs.
• Revoke credits associated with abuse, refunds, chargebacks, or policy violations.
• Limit features, rate limits, subscriptions, or payment access.
• Suspend or terminate accounts.
• Preserve records and report activity where required by law or necessary for platform safety.
• Cooperate with payment processors, hosting providers, AI providers, Waffo Pancake, or law enforcement where legally appropriate.

8. Appeals

You may appeal an enforcement decision by emailing support@textphotoai.com within 10 business days of the notice. Include your account email, the relevant prompt or output, the decision you are appealing, and a concise explanation.

We aim to review appeals within 10 business days. If an appeal is accepted, we may restore access, reverse a restriction, or correct the record. We may deny appeals involving L1 content, repeat abuse, illegal activity, payment fraud, or insufficient evidence.

9. Policy Updates and Transparency

We may update this policy to reflect product changes, legal obligations, payment processor requirements, AI provider requirements, Waffo Pancake requirements, or emerging abuse risks.

For material changes, we will provide at least 15 days' advance notice where reasonably possible. We may apply urgent safety or legal updates immediately when needed to protect users, comply with law, or prevent abuse.

10. Contact

For safety reports, content appeals, billing abuse, or questions about this policy, contact support@textphotoai.com.

Operator: LI ZHENG

Business Address: Dali Gucheng Ximen, Congyuan Village, Group 7, No. 103, Dali, Yunnan, China