Content Moderation Policy
Content Moderation Policy
Effective Date: 2025
Last Updated: 2025
z-image.ai is committed to providing a safe, lawful, and trustworthy environment for all users. This Content Moderation Policy describes how we review, filter, restrict, and manage the content submitted to or generated by z-image.ai, including text prompts, uploaded materials, and image outputs.
Our moderation rules exist to protect users, ensure responsible AI use, and comply with the requirements of Creem, international regulations, and platform partners.
1. Scope of Moderation
This policy applies to:
- Text prompts submitted by users
- Any user-uploaded images or assets (when editing features launch)
- Automatically generated image outputs
- User accounts and behavioral patterns indicative of violation
z-image.ai uses both automated moderation systems and manual review processes to enforce these rules.
2. Prohibited Content
z-image.ai does not allow the generation, storage, or processing of content in the following categories. Prompts or uploads containing any of these elements will be blocked automatically.
2.1 Sexual or Adult Content
Including but not limited to:
- Pornographic imagery
- Nudity or explicit sexual acts
- Sexualized minors (strictly illegal)
- Fetish content
- Sexually explicit roleplay or erotic scenarios
z-image.ai is a strictly NSFW-free platform.
2.2 Violence, Harm, or Abuse
Prohibited content includes:
- Graphic or gory depictions of injury
- Violent scenes involving weapons, torture, or assault
- Self-harm or suicide instructions
- Abuse, harassment, or targeted threats
2.3 Hate, Discrimination, or Extremism
We do not permit:
- Hate speech
- Extremist ideology
- Real-world political propaganda
- Targeted discrimination against protected classes
2.4 Illegal or Dangerous Activities
Including:
- Drugs or drug paraphernalia
- Weapons used in harmful contexts
- Human trafficking
- Terrorism, organized crime
- Fraud, scams, or identity deception
2.5 Intellectual Property Violations
Prohibited actions include:
- Uploading copyrighted images without authorization
- Requesting attempts to mimic the likeness of copyrighted characters
- Prompts attempting to recreate specific copyrighted works
- Attempts to generate content based on trademarked brands without permission
z-image.ai does not use copyrighted materials for training.
2.6 Deception, Misinformation, or Harmful Manipulation
z-image.ai restricts:
- Deepfake depictions of real people (celebrities, political figures, private individuals)
- Misleading or harmful synthetic media
- Attempts to impersonate businesses or institutions
- Fake credentials, badges, or endorsements
3. Moderation Methods
z-image.ai applies several layers of moderation:
3.1 Automated Prompt Filtering
Every prompt is scanned before generation. If a violation is detected, z-image.ai:
- Blocks the request
- Provides a safe alternative suggestion
- Records the event for abuse detection
No unsafe content enters the model pipeline.
3.2 Output-Level Detection
Generated images are analyzed to confirm compliance. If a violation appears in output:
- The image is immediately discarded
- The user receives a warning or block message
- Repeated violations trigger account review
3.3 Manual Review (When Needed)
For complex or ambiguous cases, our moderation team performs manual reviews.
4. User Responsibility
Users must accept that:
- They are solely responsible for the prompts they submit
- They may not attempt to bypass filters
- They may not generate content that violates this policy
- Abuse or repeated violations may result in account limitation or suspension
5. Enforcement Actions
Depending on the severity and frequency of violations, z-image.ai may:
- Block the individual prompt
- Issue a warning
- Temporarily restrict account access
- Suspend or permanently disable an account
- Report illegal content to relevant authorities (where required by law)
z-image.ai reserves the right to enforce penalties at its discretion.
6. Data Handling and Privacy
z-image.ai:
- Does not store prompts or generated images unless the user explicitly saves them
- Does not use user data for training models
- Does not retain rejected or blocked outputs
- Stores moderation logs only for anti-abuse and safety purposes
We practice minimal data retention to safeguard privacy.
7. Appeals & Support
Users may contact support@z-image.ai if they believe moderation was incorrect. We aim to respond within 24–48 business hours.
z-image.ai works continuously to refine our moderation systems to ensure fairness, safety, and compliance.
8. Policy Updates
z-image.ai may update this policy to remain compliant with evolving laws, Creem's merchant requirements, and industry standards. We encourage users to review this policy periodically.