Pre-Screening & Post-Screening Policy

At Sauci AI, operated by REGNEXT EOOD (UIC 208608387, Sofia, Bulgaria), we are committed to user safety, legal compliance, and responsible AI governance.
To maintain a secure environment for adult users, we use a dual-layer moderation system consisting of:

1. Pre-Screening – filtering and analysis before content is displayed
2. Post-Screening – monitoring and review after content is generated or published

This policy explains how these moderation layers function and how they support our Trust & Safety standards.

1. What Content is Screened

The following categories of content may be screened in both pre- and post-review phases:

• AI-generated text, images, audio, and videos
• User prompts, instructions, tags, and inputs
• Character names, descriptions, attributes, and profiles
• Usernames, bios, and account metadata
• Any uploaded or generated media through Sauci AI
• Any behavior or activity conducted on the platform

All content on Sauci AI, is subject to our safety and compliance systems.

2. Pre-Screening Review (Before Content Appears)

Before any output is displayed to the user:

• Real-time automated filters analyze prompts and uploads for violations
• Safety models detect:
  • minor-related content
  • sexual violence
  • hate speech
  • illegal activity
  • likeness misuse
  • high-risk or boundary-testing prompts
• Blocklists and pattern-recognition tools prevent known risk behavior
• In specific contexts (e.g., user-created characters), human moderators may review content before it becomes usable

Pre-Screening exists to stop harmful or illegal content from being produced in the first place.

3. Post-Screening Review (After Content Appears or Is Used)

Even after content is generated or displayed, Sauci AI continues to monitor for safety and compliance through:

Automated Post-Screening

• Continuous scanning of generated text, images, and videos
• Detection of content that may have bypassed initial filters
• System checks tied to ongoing user activity and risk patterns

Human Review & Audits

• Manual review of flagged content
• Investigation of user reports
• Ongoing audits of:
  • trending prompts
  • high-visibility content
  • borderline or ambiguous cases

User Reporting

Users may report content via: support@sauci.ai

All reports are reviewed within 7 business days, typically sooner.

Post-screening ensures that no harmful or illegal material remains on the platform, even if it initially passed through automated filters.

Users may also report concerning content through our official Discord server. While Discord is helpful for rapid communication, all formal screening-related complaints must be emailed to support@sauci.ai to ensure proper documentation and follow-up.

4. Enforcement & Escalation

If content violates our policies at either stage:

Sauci AI may take one or more actions:

• Remove or restrict the content
• Restrict access to certain features
• Issue warnings to the user
• Temporarily suspend an account
• Permanently terminate an account
• Report incidents to authorities when legally required (e.g., CSAM, trafficking, threats of violence)

Enforcement severity depends on the nature of the violation and the user’s history.

5. Commitment to Safety & Continuous Improvement

Sauci AI regularly updates its screening systems to adapt to:

• new safety threats
• emerging harmful behavior patterns
• AI model advancements
• legal and regulatory developments (including GDPR and the EU AI Act)
• user feedback and expert audits

We are committed to transparency and responsible AI operation.

6. Contact Us

For questions about this policy, to report an issue, or to submit feedback, contact: support@sauci.ai

REGNEXT EOOD
Str. George Washington 13
Sofia 1000, Bulgaria

We appreciate your help in making Sauci AI a safe and respectful environment for all adult users.