Safety & Compliance

AI Content Safety Disclaimer

NeuroTales employs multiple layers of automated safety checks to ensure age-appropriate, child-safe content. This page describes our safety mechanisms and your responsibilities as a parent or guardian.

Version 1.0.0 · Last Updated: February 2026
Automated Safety Checks Performed

Every piece of AI-generated content on NeuroTales passes through the following automated safety pipeline before being presented to users:

1

Prohibited Content Scanning

Detects violence, substances, scary content, adult themes, and bullying language across 5 pattern categories using regex-based pattern matching. Content containing prohibited terms is blocked and replaced with age-safe fallback stories.

Violence DetectionSubstance ReferencesScary ContentAdult ThemesBullying Language
2

Input Sanitization

Removes prompt injection vectors, HTML tags, code blocks, JSON-like content, and array-like patterns from all user inputs before they reach the AI model. Inputs are truncated to safe maximum lengths to prevent abuse.

3

Age-Appropriate Guardrails

Enforces vocabulary complexity, sentence length, word count, and theme restrictions calibrated to the child's developmental age band. Based on evidence from published developmental psychology research including MLU norms (Rice et al., 2010), narrative development stages (Khan et al., 2016), and DIR/Floortime FEDCs (ICDL).

4

Post-Generation Content Verification

After the AI generates a story, the full text is scanned against all prohibited patterns before delivery. If prohibited content is detected, the story is blocked entirely and replaced with a pre-written, age-appropriate fallback story. This is an enforcement layer, not just a warning.

5

Illustration Safety Filtering

Sanitizes all image generation prompts by removing scary, violent, inappropriate, and photorealistic-human terms. Enforces child-friendly art direction with bright colors, friendly characters, and children's book illustration style on every generated image.

6

Compliance Audit Logging

Records all AI interactions with safety check results for regulatory compliance. Logs include action type, input/output content summaries, safety flags, and child age — supporting COPPA (Children's Online Privacy Protection Act) and GDPR-K (EU General Data Protection Regulation for children) compliance requirements.

Known Limitations

While our automated safety system is comprehensive, no automated system can guarantee 100% accuracy or appropriateness. The following limitations apply:

  • •AI may occasionally generate content that, while passing automated checks, may not align with your family's specific values, cultural context, or sensitivities.
  • •Automated safety checks cannot detect all forms of subtle inappropriateness, cultural insensitivity, or contextual unsuitability.
  • •Image generation relies on third-party AI models with their own content policies, which may occasionally produce unexpected results.
  • •Pattern-based content detection may miss novel phrasings or context-dependent meanings that a human reviewer would catch.
Parent/Guardian Responsibility

The generating user (parent, guardian, or authorized adult) bears sole responsibility for reviewing all AI-generated content before sharing it with children or publishing it on any platform.

We strongly recommend that parents and guardians:

  • Read each story before presenting it to their child
  • Review all illustrations for age-appropriateness
  • Listen to narrations before playing them for children
  • Use the "Report Content" feature if any inappropriate material is found
  • Double-check any story before publishing to external platforms
Regulatory Compliance

COPPA Compliance

Children's Online Privacy Protection Act (USA). Parental consent required for children under 13. Data minimization principles applied. No direct data collection from children.

GDPR-K Compliance

EU General Data Protection Regulation for children. Parental consent required for children under 16. Right to erasure supported. Data processing limited to service provision.

By using NeuroTales, you acknowledge that you have read and understood this disclaimer and accept responsibility for reviewing all AI-generated content before sharing it with children or publishing it on any platform.

Despite the automated safety checks described above (Prohibited Content Scanning, Input Sanitization, Age-Appropriate Guardrails, Post-Generation Content Verification, Illustration Safety Filtering, and Compliance Audit Logging), the generating user remains the sole responsible party for the final product's safety and appropriateness.