Content Standards

When you give your bot a voice, you're shaping how it sounds to everyone who hears it. These guidelines explain what to keep in mind when writing a voice prompt or cloning a voice, and what tends to cause a voice submission to be rejected.

These articles contain our policy on bot creation: Generative AI Image Policy and the Creating & Using Bots.

Writing a voice prompt

The best voice prompts describe what the voice sounds like, not who it is. The model can build almost any vocal quality you ask for, as long as the description focuses on sound rather than identity.

Helpful things to describe:

• Pitch and timbre. Deep, light, gravelly, smooth, breathy, nasal, warm, bright.
• Pace and rhythm. Quick, deliberate, lilting, halting, measured, casual.
• Tone and mood. Confident, playful, weary, dry, earnest, theatrical, deadpan.
• Accent or region (general). A general regional class works (a Southern US drawl, a soft Irish lilt). Avoid pinning the voice to a specific person.

What to avoid in a voice prompt:

• Named real people. Voice prompts that reference real or famous people may be rejected. Cantina doesn't allow voice impersonation of real people, regardless of intent.
• Named characters. Voice prompts that reference characters from movies, TV, books, or video games are rejected.

What can cause a voice submission to be rejected

Most rejections fall into one of five patterns. Knowing the patterns helps you write a prompt that's likely to produce great results and be accepted.

1. Voices that suggest a minor

Voices that sound like children are not allowed. This includes specific age callouts (any age under 18) and descriptors like baby voice, little, young, infantilized, or childlike. This is a safety guardrail.

2. Voices of real people

Naming or clearly evoking a specific real person will be rejected. That covers politicians, musicians, athletes, influencers, public figures, your friend, or your neighbor. Real-person impersonation runs into right-of-publicity and impersonation policies regardless of intent.

3. Voices of existing characters

Named characters from movies, TV, books, video games, or other franchises are protected IP. Asking for a specific character or a thinly veiled version (changed name, same setup) violates Cantina's IP policy. Build an original character voice instead.

4. Sexualized voices

Cantina has zero-tolerance policies against the sexualization of minors and the sexualization of individuals without their consent.

Sexually suggestive voices may be rejected. Vocal qualities like breathy or whisper are fine in neutral contexts (a sleepy character, an intimate narrator). The issue is when they're paired with sexually suggestive intent.

Youth-coded language with sexual descriptors is strictly prohibited and may lead to enforcement against accounts.

5. Voices that rely on slurs or offensive caricatures

Slurs, offensive caricatures of specific ethnic groups, and explicitly racist language are rejected. Describing a general accent class for a character is fine. The issue is when the prompt relies on offensive caricatures or is explicitly racist.

Voice cloning guidelines

Voice cloning lets you make a digital copy of someone's voice for your bot. The how-to (upload, recording requirements, save) lives in How to Give Your Bot a Voice. The rules:

• Use your own voice, or one you have direct, affirmative permission to use. Default to not cloning if you're unsure.
• Don't clone a minor's voice. Child-sounding voices aren't allowed on Cantina.
• Don't clone real people without their direct, affirmative permission. If the voice belongs to someone else, you need consent from that person specifically for voice cloning or synthetic voice use. Public figures, creators, podcasters, influencers, friends, coworkers, and private individuals are all covered by the same rule.
• Don't use a cloned voice for sexual content. Cloning someone's voice and using it for sexual content is strictly prohibited and may lead to enforcement against accounts.
• When in doubt, don't clone the voice. Create an original voice prompt instead.

Before any voice clone is created, Cantina shows the creator this notice. 

Voice Clone Notice

By uploading a voice file, Cantina will create a digital clone you can apply to your AI characters.

To use this feature, please confirm the voice is yours or that you have permission to provide it, and consent to its collection, use, and storage for this purpose. Cantina will not sell your voice data, and may share it only with service providers acting on its behalf. Your voice may be used to operate and improve our products and services and will be stored securely until you delete it from your voice library.

This consent applies each time you use the feature.

If your submission is rejected

A rejection means the model flagged something in your prompt. To get a working voice:

1. Reread your prompt. Look for:

   • References to a real person (names, public figures, anyone identifiable)
   • References to a famous character (named, or a thinly-veiled stand-in)
   • Anything age-coded under 18 (specific ages, baby, little, young, childlike)
   • Sexual descriptors or suggestive vocal qualities in combination
   • Slurs, ethnic caricatures, or stereotype-based mockery

   These cover the most common triggers.

2. Replace identity with sound. Swap named people or characters for vocal qualities. "A confident, gravelly voice with a slow Southern drawl" works. "Sounds like [a specific actor]" doesn't.
3. Resubmit with the rewrite. If it still gets rejected, look for any remaining references to a real person, a famous character, a minor, sexual themes, or offensive caricatures in your prompt.

If you're stuck, the How to Give Your Bot a Voice walkthrough has examples of prompts that work.

Related Articles

• Generative AI Image Policy
• Creating & Using Bots
• How to Give Your Bot a Voice

At Cantina, we are committed to creating a safe space for all our users. Creating, sharing or distributing sexually explicit or suggestive images of other people without their consent is a severe violation of personal privacy and safety. For that reason, we take the impact of sharing intimate media of somebody very seriously and we aim to prevent the harm that the public distribution of intimate content can cause. We will take appropriate action against any content or accounts in breach of this policy.

We define the creation, sharing or distribution of non-consensual intimate media as the creation or sharing of intimate or sexually explicit content without the knowledge or consent of the person depicted. This includes media that was captured or created without the person’s awareness or consent, such as deepfakes, footage from hidden cameras, creepshots, or recordings of sexual assault. It also covers content that was initially shared or created with consent, such as in a private relationship, but later distributed without permission.

What violates this policy?

We have zero tolerance for attempts to upload unauthorized images of someone with the goal to create intimate media using our Artificial Intelligence tools. For more information, visit our Generative AI media policy.  

• Using Cantina’s tools to create sexualized content based off of somebody’s likeness.

• Attempting to manipulate or alter intimate images (i.e. deepfakes) with Cantina.

• Altering, posting, sharing, or distributing intimate images or videos of someone without their explicit consent.

• Altering, posting, sharing, or distributing private sexting photos to anybody other than the intended recipient.

• Threatening or coercing someone into sharing intimate images on the Cantina platform.

• Sharing manipulated or altered intimate images such as deepfakes without consent.

• Encouraging or facilitating others to post non-consensual nudity or sexually suggestive content.

• Attempting to obtain intimate images of someone without their consent, including hacking accounts or devices.

• Creating rooms dedicated to sharing this type of content.

• Offering or soliciting stolen intimate or sexually explicit images of somebody.

Violations of this policy may result in the removal of content, temporary account suspensions, or permanent bans, depending on the severity of the breach.

How can I report violations of this policy? 

Reporting safety violations is essential to ensuring your safety and protecting the broader Cantina community. If you encounter content on Cantina that you believe violates this policy, follow these steps to report it. In addition, if you find yourself in a room where content of this nature is being shared, report it and consider removing yourself from the space. 

If someone is sharing or threatening to share intimate images of you or someone else, you can also notify us by using this form or by emailing our Trust & Safety team on support@cantina.com. Be sure to include links and screenshots of the specific post or comment containing the images or threats.

Additional resources

In addition to reporting the content to us, there are resources available to help victims remove non-consensual intimate images, such as:

• StopNCII.org: This is a free, online tool provided by the U.K. non-profit South West Grid for Learning. It supports victims of non-consensual intimate media by helping anyone over the age of 18 protect their private images and videos online. StopNCII.org allows online platforms to detect and remove intimate media that has been shared without consent, all while safeguarding the victim's privacy. You can access the tool or learn more about how StopNCII.org works here.

• Take It Down is a free, confidential service created by the National Center for Missing & Exploited Children (NCMEC) to help individuals, particularly minors, remove or prevent the spread of intimate images or videos online. It is designed to give people control over sensitive content and to combat the harm caused by non-consensual sharing. Individuals aged 13 or older can directly submit content for protection as well as parents or guardians who can use the tool on behalf of minors under 13. For more information or to use the service, visit Take It Down.

• Support in the U.S.: If you need additional support, contact the Cyber Civil Rights Initiative Crisis Helpline at 844-878-CCRI (2274). Their team is available 24/7 to provide compassionate support, information, referrals, and non-legal advice.

• Support Outside the U.S.: For assistance in other countries, refer to the Cyber Civil Rights Initiative’s International Resources to find localized support tailored to your region.

Cantina members can browse AI-powered bots in the Bots Tab or create their own. Bots are designed to support creativity, exploration, and interaction, but they must be used responsibly and they are not a substitute for licensed professional advice, diagnosis, treatment, legal counsel, financial planning, or emergency services.

All bots and interactions are subject to Cantina's Community Guidelines. 

See also Voice Creation Guidelines.
 

Creating bots

Bots are shaped by user prompts, configurations, and design choices. If you create a bot, you are responsible for ensuring that its name, description, avatar, and prompts comply with our Community Guidelines.

You may not create bots that:

• Are or appear to be minors, including through names, bios, avatars, prompts, or behavior
• Impersonate real individuals without their express consent
• Depict or resemble public figures in a way that is misleading, deceptive, or implies an endorsement or affiliation that does not exist. Parody and satire are permitted when the bot's name, description, or behavior clearly signals it's a creative interpretation — not the real person.
• Use intentionally misleading names, descriptions, or avatars that misrepresent what the bot is designed to do

Bot creators may not design or prompt bots to generate or promote content that violates our Community Guidelines, including but not limited to:

• Hate or harassment
• Sexual exploitation or abuse
• Child endangerment
• Deepfake or deceptive content
• Violence or self-harm
• Illegal activities
• Fraud, phishing, or data theft
• Misleading or harmful advice on medical, legal, financial, or other sensitive topics

Professional advice and misrepresentation

You may not create bots that present themselves as licensed, certified, or otherwise qualified professionals when their primary purpose is to provide advice, diagnoses, treatment, legal strategy, financial guidance, or other professional services in areas where inaccurate information could cause meaningful harm.

This includes, but is not limited to, bots that represent themselves as:

• Physicians, nurses, therapists, psychologists, psychiatrists, or other healthcare professionals
• Attorneys, judges, legal advocates, or legal advisors
• Financial advisors, accountants, tax professionals, investment professionals, or similar experts

Bots may not falsely claim credentials, licensure, professional experience, or affiliations. Clearly fictional, comedic, satirical, or parody bots may be permitted when they are not reasonably likely to be mistaken for a real professional or relied upon for professional guidance.

Interacting with bots

Bot outputs are generated based on user prompts and may be inaccurate, incomplete, or inappropriate. Bots are not a substitute for professional, medical, legal, or crisis guidance, and users should not rely on them as such.

All interactions with bots must comply with Cantina's Community Guidelines and reflect authentic user engagement.

When you add a bot to a room, you are responsible for the content it generates in that space. We recommend confirming that a bot is appropriate for the room and audience before adding it. If you are unsure what a bot is designed to do, do not add it.

Prohibited bot interactions and abuse

You may not:

• Engage with bots in ways that promote hate, harassment, sexual exploitation, or violence
• Use bots to bully, intimidate, or target others
• Manipulate or misrepresent bot popularity, engagement, or ratings
• Use coordinated or deceptive tactics to artificially boost a bot's visibility

Interactions or behavior that violate these rules may result in enforcement action, including warnings, restrictions, or temporary or permanent account suspension.

Reporting violations

Our bot policies will continue to evolve as we learn more about how AI tools are used and misused. If you encounter a bot or interaction that violates these guidelines, please report it to our moderation team using in-platform tools.

If you believe your copyrighted work has been used in violation of intellectual property laws, please submit an Intellectual Property report through this form.

Cantina has a zero-tolerance policy for user actions that sexually exploit, sexualize or endanger children. We report all instances of child sexual abuse and exploitation to the National Center for Missing and Exploited Children (NCMEC) and work with law enforcement when we assess a risk of child endangerment. 

This includes any content —text, images, video, audio, or mixed media— that depicts, describes, suggests, or facilitates sexual behavior involving minors, or that sexualizes minors in any context.

This policy applies to:

• Content uploaded by users

• Content generated by users

• Content produced or enabled by Cantina’s AI tools and bots

• Prompts, prompts sets, and training artifacts used in generative systems

What violates this policy?

Cantina strictly prohibits any content, behavior, or interaction that endorses, depicts, or facilitates the sexual exploitation of children. This encompasses not only actual children, including minors of all ages but also non-realistic content that resembles human children, such as AI-generated content. The following are not permitted:

1. Sexual content involving minors

• Any depiction, description, or suggestion of a minor in sexual activity or child nudity or sexualization of minors. This prohibition applies regardless of whether the minor is real, fictional, animated, or represented through AI or generative tools.

2. Bots that sexualize minors

• You may not create, publish, promote, or engage with bots that:

  • sexualize minors;

  • simulate sexualized interaction with or about minors;

  • propagate narratives, fantasies, or descriptions that sexualize people who are or appear to be under 18.

3. Assistance in generating sexual content about minors

• Cantina tools may not be used to prompt, assist, refine, or optimize requests intended to produce sexual content involving minors.

• The platform will detect and block attempts to produce such harmful and abusive outputs.

4. Soliciting, sharing, or creating child sexual abuse material

5. Inappropriate interactions or communications between an adult and a minor

6. Any form of child abuse

How can I report violations of this policy?

If you see anything that looks like child sexual abuse or exploitation, please report it so our team can review it and take immediate action. We thoroughly review every report, whether it's from someone who's directly involved or just saw something concerning. 

Get help

If you or someone you know is being exploited, contact the Cyber Tipline at https://report.cybertip.org/ or call(1-800-THE-LOST) for 24/7 support.

Cantina allows users to create AI-generated media, including images and videos, to support creativity and expression. These tools are powerful, and with that power comes responsibility. This policy is designed to balance creative freedom with the safety, dignity, and trust of our community. 

Important notice about AI-generated media

You may encounter AI-powered bots that generate media based on user prompts in Cantina. Generated content may be inaccurate, offensive, or harmful. Do not rely on AI-generated media for factual accuracy, professional guidance, or decision-making. By creating, activating, or interacting with generative AI tools, you assume responsibility for how they are used, as outlined in our Terms of Service.

User responsibility and intent

You are responsible for the media you request, generate, upload, or use on Cantina  including media created through bots, prompts, avatars, or uploads.

Please note that while an image or video may not in and of itself violate our Generative AI policy, we reserve the right to take action if we deem that the user’s intent is in contravention of our guidelines. To do so, we may investigate factors such as account behavior, previous prompts and comments, as well as history of engagement on the platform. In addition, we hold our users accountable for the content they seek to create through Cantina’s bots, whether it is by requesting drawings, selfies, videos, designing a bot’s avatar or uploading a photo to serve as the bot’s avatar.

What violates this policy?

You may not use Cantina’s generative AI tools to create, upload, or distribute media that violates our high-priority harm policies:

1. Exploitation & sexual harm
   1. Involves child sexual abuse material, sexual exploitation, grooming, or endangerment

   2. Sexualizes minors in any form

   3. Depicts sexual violence, coercion, sextortion, or non-consensual intimate activity

   4. Includes sexually explicit deepfakes or sexualized depictions of individuals without consent

2. Harassment, hate & violence

   1. Degrades, dehumanizes, or targets individuals or groups based on protected characteristics

   2. Advocates or glorifies violence, terrorism, or extremist ideologies

   3. Encourages or provides instructions for self-harm, suicide, or eating disorders

   4. Depicts extreme or graphic violence, torture, mutilation, or animal abuse

3. Rights & likeness violations

   1. Uses someone's likeness without consent or legal right

   2. Uploads images or videos of people without permission

   3. Violates intellectual property, privacy, or publicity rights 

4. Deception, manipulation & harm

   1. Misleads others into believing AI-generated media is real or human-created in a harmful or deceptive context

   2. Creates defamatory, fraudulent, or scam-related content

   3. Generates misinformation or disinformation, including content intended to interfere with civic or political processes

   4. Advances harmful conspiracy narratives tied to real-world geopolitical events

Misleading professional representations 

You may not generate or distribute AI-generated media that falsely represents a person, bot, or fictional character as a licensed professional for the purpose of providing medical, legal, financial, or other high-risk advice. This includes media that falsely implies professional qualifications, licensure, endorsements, or expertise in a manner that could reasonably cause users to rely on the content when making important personal, legal, medical or financial decisions. 

Bot avatar guidelines  

We apply additional safeguards and review to avatar media due to the heightened risk of misuse. Bot avatars, whether uploaded or AI-generated, must comply with this policy.

• Do not upload images of real people without their consent
• Do not create avatars that depict or resemble minors
• Do not use avatars to impersonate individuals or mislead others
   

Moderation and enforcement

Cantina uses a combination of automated systems, human review, and user reports to identify and address policy violations. When violations occur, we may take action including content removal, warnings, feature restrictions, or temporary or permanent account suspension. If you are unsure whether a request or upload may violate this policy, do not proceed. 

How can I report violations of this policy?

If you encounter AI-generated media that violates this policy, please report it by flagging the content. Providing context helps our Trust & Safety team review and address issues more effectively. We also encourage users to report unintended harmful outputs as well, even when generated unintentionally.

Cantina is built to support creative exploration and meaningful connection. A safe and respectful environment is essential to that mission. For this reason, bullying and harassment are not allowed on Cantina.

We take action against behavior that targets, intimidates, or harms others, whether it occurs in public spaces, private interactions, profiles, room names, or through AI-powered features. Users who violate this policy may be subject to warnings, restrictions, or temporary or permanent account suspension.

What violates this policy?

We prohibit any act, whether on a public surface such as room posts, avatars or account names, or a private engagement, to harm, intimidate, coerce, or embarrass another person or group. Examples of this type of conduct include but are not limited to:

• Repeated or severe abusive language directed at or about another user
• Content intended to shame, degrade, or humiliate others
• Repeated unwanted interactions, including following a user across rooms
• Coordinated harassment, including encouraging others to target a user
• Harassment that extends off-platform when it originates on Cantina
• Filing false or malicious reports against another user
• Impersonation or creating profiles designed to mock or harass someone
• Threats of physical, emotional, reputational, or social harm
• Encouraging or calling for self-harm or suicide
• Threatening to expose private or intimate information, including sexual content or a person’s sexual orientation or gender identity
• Creating or renaming rooms, profiles, or avatars in ways that are intentionally offensive or targeted at others

AI and Harassment

Harassment is not permitted whether carried out directly by a user or facilitated through bots or AI tools. Using automated systems to target, intimidate, impersonate, or coordinate harassment against others violates this policy.

How can I report violations of this policy?

Reporting content and interactions that violate our Community Guidelines is essential to fostering a safe environment in Cantina. You can report conduct in violation of this policy by following these instructions. In addition, we encourage you to use our moderation tools to keep yourself and others safe.

If you feel like you or another user is in immediate danger, we encourage you to consider contacting Law Enforcement urgently.

Cantina is committed to fostering a respectful, inclusive environment where people can create, connext, and express themselves without fear of being targeted or dehumanized. Hate speech and hateful conduct are not allowed on Cantina. 

We take action against content or behavior that promotes hatred, discrimination, or violence toward others based on protected characteristics. Violations of this policy may result in content removal, account restrictions, or temporary or permanent suspension.

What violates this policy?

We do not allow content or conduct that promotes, glorifies, or incites hatred, discrimination, or violence against an individual or group based on a protected characteristic, including but not limited to:

• Age
• Caste
• Disability
• Ethnicity
• Gender identity or expression
• Nationality
• Race
• Immigration status
• Religion
• Sex or gender
• Sexual orientation
• Veteran status
• Victims of major violent events and their families

This includes:

• Calls for violence, removal of rights, or exclusion
• Dehumanizing, degrading, or inflammatory language
• Praising or advocating for extremist or hateful ideologies
• Harassment or intimidation rooted in a protected characteristic

This policy applies across all surfaces on Cantina, including rooms, profiles, messages, usernames, room names, and interactions involving bots or AI-powered features.

Context matters

We recognize that not all references to hateful language or symbols are themselves hateful.

The following may be allowed when the context is clear and non-promotional:

• Educational, documentary, scientific, journalistic, or artistic content that references hate speech, symbols, or ideologies for the purpose of reporting, analysis, or critique
• Content criticizing governments, institutions, policies, or state actions, even when strongly worded, as long as it does not target people or groups based on protected characteristics
  
  We assess content based on context and impact, not solely stated intent. Claims of humor, satire, or roleplay do not excuse hateful conduct when it causes harm.

AI and Hate Speech

Hate speech is prohibited whether it is expressed directly by a user or generated, amplified, or facilitated through bots or AI tools. Using AI to produce or spread hateful content, or to target protected groups, violates this policy.

Reporting violations

If you experience or observe hate speech on Cantina, please report it so our team can review it and take appropriate action. We take all reports of hate speech seriously, whether they come directly from the person being attacked or from an observer. You can also use our self-moderation tools to manage room membership and limit your exposure to the offending user and/or content. 

Cantina does not allow violence, threats of harm, or behavior that creates a credible risk to people or public safety. This includes direct threats, incitement, or depictions intended to glorify or encourage real-world harm.

Violations of this policy are treated as serious and may result in immediate content removal, account restrictions, suspension, or termination.

If you believe you or someone else is in imminent danger, please contact local emergency services or law enforcement immediately. 

What violates this policy?

We have a zero-tolerance policy for content or conduct that threatens, incites, or depicts real-world harm, including but not limited to:

• Threats of physical harm or violence against individuals or groups
• Calls to commit or encourage violent acts
• Depictions or glorification of high-severity violence, abuse, or extreme gore
• Content that meaningfully facilitates or endorses real-world violent wrongdoing
• Harassment or intimidation that escalates into credible threats to safety

This policy applies across all areas of Cantina, including rooms, profiles, messages, usernames, room names, and interactions involving bots or AI-powered features.

When we assess a credible risk of physical harm or threats to public safety, we may take immediate action and work with law enforcement as appropriate.

Context

We recognize that some content may reference violence in educational, documentary, historical, or artistic contexts. Such content may be permitted when the context is clear and does not promote or encourage harm.

We assess behavior based on context, severity, and potential impact, not solely stated intent.

Reporting violations 

If you experience or observe threats of harm, violence, or illegal activities, please report it so our team can review it and take appropriate action. We take all reports seriously, whether directly from the person being attacked or an observer. You can also use our self-moderation tools to manage problematic behaviors in your rooms.

Cantina is committed to creating a safe and welcoming environment. We support creative and expressive interactions between adults but only when they are clearly consensual and context-appropriate.

We strictly prohibit sexual behavior or content that is non-consensual, coercive, exploitative, or intended to shock, humiliate, intimidate, or violate others, as well as nudity or sexually explicit material in prohibited contexts. Violations of these policies may result in account restrictions, suspension, or termination.

What violates this policy?

You may not create, share, or engage in sexual behavior or content that:

• exploits, harms, coerces, or harasses others
• involves unwanted sexual attention, pressure, or solicitation
• exposes others to sexual content without their consent
• occurs in public or mixed-audience spaces where it is not appropriate
• targets Cantina staff or representatives with sexual requests
• uses bots or AI tools to generate or facilitate non-consensual or exploitative sexual content

Sexual content involving individuals under 18 is strictly prohibited. Cantina reports such cases to NCMEC and cooperates with law enforcement. In addition, public or mixed-audience spaces must not feature live streams, photos, videos, or AI-generated content depicting nudity, sexual acts, simulated sexual acts, or sexually explicit or unavoidable sexual imagery, especially when it is non-consensual or intended to shock or degrade others. 

These rules apply equally to human-generated and AI-generated content, including chatbots, image tools, and other generative features.

What is not a violation of this policy?

We recognize that artistic, educational, documentary, or scientific content may sometimes include nudity or references to sexuality. We may allow this content when it is clearly contextual, non-sexualized, and does not violate our Community Guidelines.

We reserve the right to remove any content we determine to be inappropriate, unsafe, or inconsistent with the spirit of this policy.

How can I report violations of this policy?

If you experience or observe sexual behavior or content that feels non-consensual, exploitative, or unsafe, please report it using our in-platform tools. We take all reports seriously, whether they come from the person affected or an observer, and we encourage the use of self-moderation tools to manage inappropriate behavior in live or shared spaces. 

If you or someone you know is being exploited, contact the Cyber Tipline at https://report.cybertip.org/ or call (1-800-THE-LOST) for 24/7 support.