This Acceptable Use Policy (“AUP”) governs what you may and may not do with Sidecaster (the “Service”), operated by PEEK, LLC. It is part of our Terms of Service. Because Sidecaster produces text, synthesized voice, and images that you broadcast to a live audience, these rules protect you, your viewers, third parties, and the Service. Violating this AUP may result in suspension or termination without refund.
1. Prohibited content and conduct
These prohibitions track the acceptable-use standards of the AI providers we build on (including OpenAI, Anthropic, ElevenLabs, and Inworld) and flow down to your use of Sidecaster. You may not use the Service to create, request, distribute, or broadcast content in any of the following categories:
- Children’s safety.Child sexual abuse material (“CSAM”), including AI-generated CSAM; or content that fetishizes or sexualizes minors, including in fictional settings or via roleplay.
- Sexually explicit content. Sexual intercourse or sex acts, pornographic or obscene material, erotic chat, or content related to sexual fetishes or fantasies.
- Self-harm. Content that facilitates, promotes, or glamorizes any form of suicide or self-harm, including eating disorders, disordered eating, and unhealthy or compulsive exercise.
- Violence and hate. Content that incites, facilitates, or promotes violent extremism, terrorism, or hateful behavior, or that promotes discriminatory practices against individuals or groups on the basis of protected attributes.
- Graphic violence. Content that promotes, trivializes, or depicts graphic violence or gratuitous gore, including sexual violence.
- Harassment. Content that shames, humiliates, intimidates, threatens, bullies, or harasses an individual, or that coordinates the harassment or intimidation of a person or group.
- Impersonation and deception. Impersonating a human by presenting AI output as human-generated, or otherwise convincing people they are interacting with a person when they are not (see also the mandatory AI disclosure requirement in Section 7); or impersonating a business, brand, or organization in a deceptive or harmful way.
- Voice and likeness.Replicating another person’s voice or likeness, in any form or for any purpose. (Sidecasters may not be configured as real people at all — see Section 2.)
- Privacy. Violating privacy rights, such as sharing personal information without consent, or collecting or exposing private information such as non-public contact details, health data, or biometric data.
- Intellectual property. Infringing, misappropriating, or violating the intellectual property rights (such as copyright or trademark) of a third party.
- Misinformation.Deceptive or misleading information targeting a person, group, or entity — including defamation (false statements of fact that harm a person’s reputation) — or false or misleading information related to medical, health, or science issues.
- Illegal or regulated goods. Acquiring, exchanging, or facilitating the acquisition of illegal or controlled substances, weapons, or other unlawful goods or services, or facilitating human trafficking.
- Unqualified professional advice. Output presented as professional financial, legal, or medical advice without qualified human review and disclosure that the content is AI-generated.
- Fraud and scams. Content for fraud, scams, phishing, or spam, or falsified documents such as fake IDs.
- Systems and infrastructure. Malware, unauthorized access to systems, or content that facilitates the disruption of critical infrastructure such as power grids, water systems, or telecommunications networks.
2. Real people — no impersonation
You may not configure a sidecaster to impersonate or replicate any identifiable real person, living or deceased. This includes a person’s name, likeness, voice characteristics, mannerisms, catchphrases, or distinctive biographical details. This rule applies regardless of whether you claim to have the person’s consent.
3. Political figures — absolute prohibition
You may not configure a sidecaster to represent any political candidate or elected or appointed government official, regardless of whether authorization was obtained. You also may not, in any case and regardless of authorization, use the Service to create synthetic media of political figures, or content designed to deceive or mislead voters or to suppress or interfere with electoral or civic processes. These prohibitions are absolute.
4. Third-party fictional characters
You may not configure a sidecaster to portray a specific named character owned by someone else (for example, characters from Disney, Marvel, Lucasfilm, Nintendo, or similar). Generic archetypes are fine — “a snarky pirate” is allowed; a specific named, protected character is not.
5. Minors
You may not configure a sidecaster that is, or is depicted as, a minor. This is stricter than the general CSAM prohibition and applies to all depictions of children.
6. Voices and voice labeling
Stock voices provided by the Service may not be labeled, described, or framed to evoke or imply a specific real person. You may not attempt to present a generic voice as the voice of a particular individual.
7. Mandatory AI disclosure to your audience
When you use Sidecaster on a live stream, you must clearly and conspicuously disclose to your audience that an AI co-host is in use, so that a reasonable viewer is not misled into believing the co-host is a human. This is required by us, by our upstream AI providers, and by a growing body of law governing AI disclosure.
You choose how to disclose, so long as the disclosure is clear and conspicuous. Acceptable methods include noting the AI co-host in your stream title or category, stating it in a channel panel, bio, or “about” section, and/or saying so on stream. You must not configure the co-host to deny that it is AI or to claim to be a specific real person (see Section 1), and you are responsible for keeping your chosen disclosure accurate and reasonably visible to viewers who could otherwise be misled.
8. Twitch and other platform rules
You represent that your use of the Service complies with the terms of service of Twitch and any other platform you broadcast to. You are solely responsible for any enforcement action a platform takes against your channel in connection with your use of the Service.
9. Enforcement
We enforce this AUP through a combination of measures: the safety systems of the AI providers we build on; automated checks that screen certain requests — such as image-generation prompts — for high-risk categories like child sexual abuse material before they are processed; logging of content-safety decisions; and review of reports we receive from users and third parties. We do not pre-screen all Output (see our Terms of Service), and you remain responsible for the content you generate and broadcast.
When we identify a violation, we may remove content and suspend or terminate accounts, with or without notice. We report apparent child sexual abuse material to the National Center for Missing & Exploited Children (NCMEC) as required by law, and may report other unlawful content to the appropriate authorities. To report a violation or a specific piece of content, contact legal@peekgames.dev.