Outspeak

AI Safety at Outspeak

We are committed to building AI tools that are safe, responsible, and respectful of people and communities. This document outlines how we approach safety on Outspeak, what content is disallowed, and how we enforce our policies.

Disallowed content

You may not use Outspeak to create, generate, or distribute content that includes:

  • Creating political disinformation or propaganda
  • Creating content for fraudulent purposes or to promote scams
  • Creating hate speech or discriminatory content
  • Harassment, bullying, or threats of violence
  • Explicit sexual content involving minors or non-consensual activity
  • Incitement to violence, terrorism, or criminal activity
  • Biometric spoofing or attempts to impersonate an individual without clear, informed consent
  • Content that violates intellectual property rights or privacy rights

We may update or clarify these examples over time. When in doubt, please contact us before generating content that could be sensitive or high‑risk.

How we promote safe and responsible use

We use a combination of technical and operational measures to reduce misuse and harmful outcomes:

  • Automated tools to filter content before it is generated
  • Storing metadata and select samples of audio and video generated through our service
  • Content fingerprinting techniques
  • Abuse monitoring and rate limiting
  • Policy‑informed models and safety classifiers
  • Human review for escalations and repeat abuse

We continuously improve our protections and welcome feedback from our users and the broader community.

Transparency and labeling requirements

We require our users to clearly label any content they share on social media platforms like Instagram and TikTok as “AI-generated” or with a similar label, whenever such labeling options are available. We encourage you to provide additional context where appropriate (for example, that a video uses an AI voice or AI lip‑sync).

Consent and impersonation

When creating content using someone’s likeness or voice, you must obtain clear, informed consent. Content that attempts to deceive audiences about a person’s actions, beliefs, or endorsements is prohibited.

Enforcement

We reserve the right to monitor, remove, or disable access to any content that we determine to be in violation of our policies. We may also terminate or suspend accounts that engage in misuse of our service, and no refunds will be issued in such cases.

If you believe content on our platform violates these policies, please report it to us at contact@outspeak.ai.