Heartbits AISubscribe

HeartBits Content Safety and Review Policy

Effective Date: July 13, 2026
Last Updated: July 13, 2026

1. Purpose

HeartBits (the "Platform") is committed to providing safe, lawful, and trustworthy AI content generation services. To prevent unlawful, harmful, or rights-infringing content from being generated or distributed through the Platform, HeartBits maintains post-generation human review, prohibited-term and prompt management, review-record retention, and enforcement procedures.

This Policy applies to text prompts, character settings, images, videos, audio, and other materials submitted through HeartBits, as well as images, videos, and related content generated from those materials.

2. Content Review Stages

2.1 Post-Generation Human Review

After an AI image or video has been generated, the Platform conducts a human content safety review of the result. The content may remain in a processing or pending-review status until the review is complete.

Human reviewers assess the user's input, the generated result, the intended use, and applicable Platform policies. Content that passes review may be displayed or delivered to the user. Content that does not pass review will be blocked and will not be displayed or delivered.

2.2 Ongoing Re-Review and Enforcement

Content that has already been displayed, delivered, or published may be reviewed again if the Platform receives a user report, rights-holder complaint, regulatory notice, or other information indicating that the content may violate applicable law or Platform policies. The Platform may take additional action based on the outcome of that review.

3. Scope of Review

The Platform may review:

  1. Text prompts, character settings, and generation instructions submitted by users;
  2. Images, videos, and other materials uploaded or selected for generation;
  3. Images, videos, and related content produced by AI models;
  4. Generated content that users publish, share, or distribute through the Platform; and
  5. Account activity, generation frequency, and relevant history associated with suspected violations.

4. Review Standards

Reviews are conducted under applicable laws and regulations, regulatory requirements, the HeartBits User Agreement, Privacy Policy, and Prohibited Content Policy. Content may be rejected, blocked, or removed if it involves:

  • Illegal activity, violent extremism, terrorism, or dangerous conduct;
  • Sexually explicit content, sexual exploitation, or any inappropriate content involving minors;
  • Hate, discrimination, harassment, bullying, or threats;
  • Fraud, impersonation, false advertising, or material deception;
  • Infringement of copyright, trademarks, likeness rights, privacy rights, or other third-party rights;
  • Disclosure of personal data, trade secrets, credentials, or other non-public information;
  • Illegal drugs, gambling, unlawful transactions, weapons, or restricted goods and services;
  • Attempts to evade review, bypass safety measures, or induce a model to violate Platform rules; or
  • Other content that violates applicable law, public order, or Platform policies.

5. Review Outcomes and Enforcement Measures

Depending on the nature and severity of a violation, its impact, and the user's history, the Platform may take one or more of the following actions:

  1. Reject the generation request;
  2. Mark the generated result as not approved;
  3. Prevent the content from being displayed, downloaded, shared, or published;
  4. Remove or disable access to the content;
  5. Restrict access to some or all generation features;
  6. Suspend or terminate the associated account;
  7. Preserve necessary records and cooperate with lawful investigations by competent authorities; or
  8. Take any other action permitted or required by applicable law.

6. Prohibited-Term and Prompt Management

The Platform maintains a prohibited-term and prompt-filtering library to identify high-risk generation requests, support human review, and reduce the risk of prohibited content being generated.

The library covers illegal activity, sexual content and child safety, violence and terrorism, hate and harassment, fraud and impersonation, privacy violations, intellectual property infringement, dangerous goods, illegal drugs, gambling, cyber abuse, and review-circumvention attempts.

The Platform performs a routine update of the library once per month. Emergency updates may also be issued in response to changes in law, regulatory requirements, risk events, user reports, or review findings.

7. Retention of Review Records

The Platform retains necessary content review records for content safety management, regulatory review, dispute resolution, and security audits.

Review records may include a user or account identifier, request time, content type, necessary input information, generated-result identifier, review time, review decision, violation category, and enforcement outcome.

Review records are retained for 180 days from the date they are created. If a longer retention period is required by law, regulation, dispute-resolution needs, or a lawful request from a competent authority, the Platform will retain the relevant records as required. When the applicable retention period expires, records will be deleted or anonymized unless continued retention is required by law.

8. Appeals and Reports

Users who believe that content was incorrectly blocked or that an account was improperly restricted may submit an appeal by email. Appeals should include the relevant account information, content identifier, generation time, and reason for the appeal. The Platform will review the submission and respond after verification.

Users may also report content that may be unlawful, infringing, or otherwise unsafe. The Platform will review the submitted information and take appropriate action.

9. Policy Updates

The Platform may update this Policy to reflect changes in law, regulatory policy, model capabilities, or risk-management practices. An updated version will be published on the Platform and will apply from its stated effective date.

10. Contact Us

For questions about content review, enforcement, reports, or appeals, please contact: