Responsible AI

Frameworks

Policies, guidelines, and how Skytells operationalizes responsible AI.

How We Operationalize Responsible AI

Skytells turns its commitment to responsible AI into concrete policies, guidelines, and processes. This page summarizes the frameworks we use and where to find more detail.


Policy and Guidelines

ElementDescription
Tiered safety frameworkAll models (Skytells and partner) are under Skytells safety guidelines with moderate moderation by default (Tier 1). Super-realistic models use Tier 3. See Safety Tiers.
Safety guidelinesRules and guardrails for all models; enhanced for high-fidelity models (e.g., TrueFusion, Mera). Define acceptable use, content filters, and when prior approval is required. See Safety.
No-cloning ruleYou may not create a likeness or clone of a real, identifiable person without consent and an authorized use case. Covers face, voice, and identity. See Responsible AI — Output Moderation.
Default output moderationOutputs are moderated by default to reduce DeepFake risk and other misuse. Some content may be blocked or require approval.
Ethics and fairnessWe apply fairness and bias mitigation in design and evaluation, and publish resources on ethics, fairness, and bias. See Ethics.

Prior Approval and Access Control

  • Some features or models require prior approval for safety reasons. Access may be gated by use case, account type, or compliance review.
  • Approval requests are handled through Skytells support or your account team. Provide a clear description of your use case and how you will comply with our policies.
  • We may update which capabilities require approval as risks and regulations evolve.

External Resources and Compliance

We align our practices with widely used norms and, where relevant, regulatory expectations:

Our Security and Privacy docs describe how we protect infrastructure and personal data; responsible AI frameworks complement these with content and use-case safety.


Staying Updated

Policies and frameworks may change as we respond to new risks, research, and regulation. Important updates are communicated via:

For questions about how these frameworks apply to your use case, contact Skytells support.

How is this guide?

On this page