AI Content Guidelines

Introduction and Scope

Fantra is an adult-only (18+) platform featuring AI-generated fictional content (all created by users off-platform). These AI Content Guidelines apply to all AI-generated content on Fantra, whether produced with external AI tools by creators or (in the future) via any AI features Fantra might introduce. The guidelines ensure that all AI-created works shared on Fantra remain legal, ethical, and in line with global compliance standards. (If there is any conflict with other policies, these Guidelines govern AI-specific matters.) Violations may lead to enforcement action as described below

Scope: These Guidelines cover all forms of AI-generated content on Fantra (images, videos, text, audio, etc.), as well as the prompts/inputs used to generate such content. They apply equally to content produced on the Fantra platform and to any AI content you might upload (subject to Fantra’s rules permitting such uploads). All content must comply regardless of creation method.

Definitions

For clarity, here are key definitions related to this document.

• AI Generated Media means any image, video, audio or text created or materially altered using artificial intelligence

• Prompt means the instructions or inputs a user provides to generate content, Output means the result produced by the AI

• Deepfake means AI Generated Media that depicts a real identifiable person without their written consent

• Fictional Characters Only means that all depicted persons must be fictional adults and not real people

• Bypassing Filters means any attempt to evade safety systems or use coded language to generate disallowed content

• Extreme Content means material that a reasonable person would find highly obscene or deeply disturbing and that is not permitted under these Guidelines

Acceptable Use of AI-Generated Content

Fantra allows users to generate and share a wide range of adult-oriented fictional content using AI. Creative expression is encouraged, including erotic or fantasy scenarios, provided they involve consenting adults and comply with these rules. Permissible AI-generated content can include realistic or stylised depictions of fictional characters in various genres (e.g. romance, sci-fi, fantasy) and artistic styles (from photorealistic to anime or cartoon). The key requirements are that all depicted characters are adults, the content is consensual, and no laws or platform policies are broken.

• Fictional Characters Only: All characters in your AI-generated content must be fictional creations. You may not depict real persons (living or deceased) unless you have their explicit written consent and they are 18 or older. Fantra is an AI-only content platform, so no actual photos/videos of real people are permitted. This ensures creativity is focused on fictional scenarios and protects individual rights and privacy.

• Adult Themes with Consent: Sexual or erotic AI content is allowed only between consenting fictional adults. Fantra’s core content is adult-oriented, so you may explore various adult themes, kinks, and fantasies within consensual and legal boundaries. Always ensure that any erotic scenario clearly involves adults who can consent and avoid any suggestion of minors or non-consent (more on these prohibitions below).

• Creative Freedom and Style: You have flexibility in art style and narrative tone. AI-generated content may be realistic, fantastical, or stylised (e.g. anime-style, cartoon, CGI). Feel free to experiment with genres and aesthetics. However, no matter the style, the depicted content must not violate these Guidelines. For example, a cartoon image is not exempt from rules against illegal or abusive themes. Styling (e.g. making a character look like an elf or an alien) is fine as long as it adheres to content rules – e.g. an adult alien character is allowed, but a “child-like” cartoon character is not.

• Transparency: While Fantra’s nature as an AI platform is generally understood by users, you should not attempt to pass off AI-generated content as real to mislead others. If you share Fantra-generated content outside the platform, it’s good practice to clarify its AI-created. Fantra may also require labelling of AI content in certain contexts (e.g. a watermark or caption indicating AI origin), and you must comply with such requirements.

In summary, use AI responsibly and imaginatively. Stick to fictional, adult scenarios, respect the consent of all depicted characters, and stay within legal bounds. If you are ever unsure whether certain content is allowed, err on the side of caution or seek guidance from Fantra support before proceeding.

Strictly Prohibited AI-Generated Content

Fantra maintains zero-tolerance for certain content categories. Some themes are strictly forbidden even in AI-generated, fictional form. Generating or sharing the following will result in immediate removal and likely account suspension or termination:

• Any Depiction of Minors: Content that involves or appears to involve minors (under 18) in any sexual, violent, or exploitative situation is absolutely banned. This prohibition includes:

  • Actual Minors: You may not create an AI likeness of any real person under 18 (e.g. using a child’s photo or a famous minor’s name).

  • Fictional Underage Characters: You may not fictionally portray anyone under 18 in a sexual, abusive, or otherwise adult context. All characters in erotic or violent content must be and appear to be adults. Any form of “age-play” scenario or attempts to make a character seem underage (via youthful appearance, school settings, underage role labels, childish mannerisms, etc.), is strictly forbidden. Even in fantasy or cartoon style, child-like depictions are prohibited. We have zero tolerance for child sexual abuse material (real or AI-generated); any attempt to create such content will lead to immediate account termination and reporting to authorities.

• Non-Consensual or Exploitative Sexual Content: Rape, sexual assault, or any non-consensual sexual scenario is banned outright – even if portrayed fictionally or “just fantasy.” All sexual content must be consensual. You may not depict sexual violence, coercion, or exploitation (e.g. rape fantasies, someone being drugged or forced, sex trafficking scenarios) . This includes incest (sexual content between family members) which is also disallowed on Fantra. Consent is paramount – if a storyline involves someone saying “no” or any form of unwillingness, it crosses the line. Likewise, any depiction of individuals who cannot consent(unconscious, extremely intoxicated, etc.) is forbidden. Fantra does not allow content that eroticises lack of consent or powerlessness.

• “Child-Like” Sexualised Characters: To reinforce the above points – you cannot use the AI to create adult characters that look underage. For example, an 18+ character with a very child-like body or dress (like a pre-pubescent appearance, toddler-like speech, or wearing a school uniform implying a minor) is not allowed. If an AI-generated person could be reasonably mistaken for a child or early teen, do not use them in sexual content. When in doubt, adjust the character to appear clearly adult (older features, mature attributes). Moderators use a “reasonable person” standard to judge this – any content that even appears to sexualise children will be removed.

• Real People Without Consent (Deepfakes/Impersonation): You must not use Fantra to create deepfake or AI-generated content that depicts real individuals (whether celebrities or private persons) in any explicit or derogatory scenario without **written consent from that person. Using a real person’s likeness in pornographic or sexual AI content is a serious violation of privacy and consent. This includes:

  • Celebrities/Public Figures: Do not create nude or sexual images of famous people using AI (or any realistic lookalikes). Even if the person is an adult, if you don’t have their permission, it’s banned.

  • Private Individuals: Do not upload a photo of an acquaintance or any person to “transform” it into sexual content. Fantra does not permit any user-uploaded imagery of real humans at all. Attempting to disguise a real photo as AI output is a violation. Also, you cannot write text prompts that identify a real person and generate sexual content about them.

  • Impersonation and Roleplay of Real People: You may not even fictionally roleplay as a real, identifiable person in an erotic or defamatory way on Fantra. For example, writing a scenario using a celebrity’s name or another user’s persona in sexual context is disallowed. Any content that sexualises or harasses a real person without consent (even in text) is forbidden. Fantra’s content is about fictional characters – keep real individuals out of it unless explicitly authorised.

• Bestiality and Animals: No sexual content involving animals. This includes real animals or realistic AI-created creatures that are essentially animals. Sexual acts with animals (bestiality) or animal-like creatures are prohibited. Exception: Fictional fantasy beings that are intelligent and consenting (e.g. a mythical creature or alien with human-like sentience) may be permitted if clearly fictional and adult. But use extreme caution – anything that resembles real animal abuse or non-consensual creature scenarios is banned. When in doubt, avoid the scenario.

• Extreme or Obscene Material: Even with consenting adults, certain extreme content is not allowed. Do not generate content that a reasonable person would find highly obscene or deeply disturbing, especially if it violates law. This includes, but isn’t limited to: sexual torture, snuff or murder fantasies, necrophilia (sex with corpses), cannibalism, extreme gore associated with sexual context, or scatological fetishes (e.g. consumption of human waste) . Any fetish or scenario that involves severe harm, risk to life, or portrayal of serious physical abuse is forbidden, even if all characters are fictional adults. Keep content within the realm of legal, consensual fantasy – extreme violence or cruelty for shock value is not welcome on Fantra.

• Illegal or Harmful Acts: Do not use AI to produce content that facilitates crime or depicts illegal acts in a promotional way. Generating how-to guides for making weapons, depictions of criminal activity intended to encourage replication, or content that glorifies real-world violence, self-harm, or drug abuse is disallowed. Non-sexual violent content (like fantasy battle scenes) is generally permissible in moderation, but extreme real-world violence, terrorism propaganda, or graphic crime scenes are not. Also, content that encourages suicide, self-harm, or eating disorders is forbidden (you may depict mental health struggles in a respectful way but never glorify or encourage self-harm) .

• Bypassing Filters or Evasion Attempts: You must not attempt to trick or evade Fantra’s content moderation filters. This includes using code words, alternate spellings, or obscure references to generate disallowed content. For example, don’t try to get around the “no minors” rule by using coded language or slang for underage characters. Any deliberate attempt to circumvent the AI’s safeguards (like repeatedly tweaking a prompt that was blocked for policy reasons) is a violation on its own. Do not use external tools or hacks to manipulate Fantra’s AI into breaking these rules. Such behaviour may lead to immediate suspension. If the AI system flags your content as disallowed, take that as final – do not persist in trying to game the system.

This list of prohibited content is not exhaustive. Any content that is illegal, highly unethical, or which Fantra deems contrary to its mission of a safe adult community is disallowed. When in doubt, err on the side of caution. Fantra’s moderators may remove content that, even if not explicitly listed above, is judged to pose legal or safety risks. Always keep your AI creations within the bounds of law and decency.

Consent and Power Dynamic Boundaries

All sexual and adult content on Fantra must depict consenting adults. Consent isn’t just about “yes or no” – it must be free of coercion, manipulation, or undue power imbalance. When crafting AI-generated scenarios, be mindful of power dynamics between characters:

• No Non-Consensual Narratives: As stated, rape or sexual violence is banned. Beyond outright force, avoid any scenario where consent is unclear or pressured. For instance, a storyline where one character is an authority figure (teacher, boss, doctor, etc.) and uses that power to obtain sex from a subordinate could verge into non-consensual territory. Such content is strongly discouraged or forbidden unless it’s abundantly clear in the narrative that it is a consensual role-play scenario with no real coercion. Even then, proceed with caution – if a scenario too closely mirrors real-world exploitation (e.g. a prisoner and guard), it will likely be disallowed.

• Power Imbalances: Fantasies involving power differences (like master/servant, teacher/adult student, etc.) must be handled responsibly. If you explore these, both parties still need agency and consent in the depiction. Under no circumstances can one party be truly incapable of refusing. For example, “hypnosis” or mind-control erotica is effectively non-consensual and thus not allowed if it removes a character’s ability to consent. Likewise, blackmail scenarios (sexual extortion) are forbidden. All roleplay power dynamics must be explicitly consensual and negotiated within the story.

• No Exploitation of Vulnerability: Do not depict sexual scenarios that take advantage of someone’s desperation, incapacitation, or innocence in a predatory way. For instance, an authority figure trading necessities (money, shelter) for sex from someone in dire straits can be deemed exploitative content and banned. Characters portrayed as significantly unable to make free choices (due to drugs, mental state, or inequality) cannot be used in sexual content.

• Incest and Familial Roles: As noted, incest content is disallowed. Even among consenting adults, Fantra does not permit sexual content between family members or close relatives, nor “roleplay” of such scenarios. Avoid any depiction of parent/child, siblings, or similar relationships in an erotic context – it’s not only non-consensual by nature (if a parent/child, one is a minor) but also against our standards even if you claim both are “adults.” Family roleplay is off-limits.

The guiding principle is respect and realism in consent. You are free to create erotic fantasies, but they should not glorify real abuses of power or non-consensual acts. When both characters are consenting adults on equal footing (even if one is play-acting a submissive role), that’s within bounds. The moment a scenario replicates real-life harm or coercion; it crosses into prohibited territory. When writing prompts or stories, consider: would all participants genuinely agree to this in real life? If not, don’t generate it.

Realism, Fantasy, and Style Guidelines

AI art and content can range from cartoonishly unreal to hyper-realistic. Fantra welcomes different artistic styles, but with these considerations to keep content within our rules:

• Hyper-Realistic Images: If you create very realistic AI images, remember that all our rules still apply. A lifelike image should not be mistaken for a real person who hasn’t consented. Ensure your realistic characters are entirely fictional and do not resemble an actual person (unless you have consent as noted). With photorealism, also be extra cautious that your characters look clearly adult – sometimes a realistic AI face can appear young inadvertently. Always adjust prompts or use tools to make sure any realistic depiction appears 18+. If an image is so realistic that viewers might think Fantra is hosting real pornography, we may require labelling it as AI-generated for transparency.

• Fantasy and Mythical Content: You are allowed to generate fantasy scenarios – e.g. fictional creatures, aliens, vampires, magical settings – as long as they don’t violate content rules. Fantasy violence (like a battle with a dragon) is acceptable unless it becomes excessively graphic. Sexual content with fantastical beings is permitted only if those beings are depicted as sentient adults capable of consent (see the bestiality rule: no real animals, but a consenting humanoid mythical creature could be okay) . Remember that even fantasy realms aren’t an excuse for illegal themes – e.g. “500-year-old demon in the form of a child” is absolutely not allowed. The form that appears underage is what matters, not technical age in lore.

• Stylised or Cartoon Forms: Cartoon or anime-style content is popular and allowed. However, a cartoon depiction must still follow all guidelines. Sometimes, cartoon styles can inadvertently make characters look young – avoid overly childlike proportions (like very small stature with childlike faces) in adult content. If you use “chibi” or highly stylised forms, ensure any sexual context is clearly adult and the style doesn’t mimic underage imagery. Also, extremely violent or graphic acts can be jarring even in cartoon form and are disallowed just as they would be in realistic form. Stylised violence (like comic-book action) is tolerated, but not if it involves sexual assault or extreme gore.

• Quality and Clarity: While not a strict “legal” rule, we encourage creators to strive for quality and clarity in their AI outputs. The clearer your content, the less likely it will be misinterpreted. Blurry or distorted AI images that could be mistaken for something disallowed (e.g. an AI image glitch that makes a character look too young) should be fixed or avoided. Use prompts that clearly define adult, consenting context. If you generate story text, avoid ambiguous descriptions that might imply non-consent or underage participants without explicitly meaning to. Essentially, be explicit in keeping things compliant – it helps our moderators and ensures your content stays up.

• No Misleading Uses of Style: Do not use filters or styles to try and hide policy violations. For example, making an image black-and-white or in a painting style does not make a forbidden theme acceptable. (One cannot say “it’s just art” to justify illegal content.) Similarly, don’t think that because something is obviously fantasy (e.g. set in a fictional world) that rules on consent or minors don’t apply – they do. Moderation will look through the style to the actual substance of the content.

AI Prompting and Monitoring

User Prompts and AI Inputs: If you use an AI tool in connection with Fantra (for example, an external generator to create content you plan to upload, or any future Fantra-provided AI feature), any text prompts or commands you use must adhere to these Guidelines. In other words, you must not create or attempt to create content through AI prompts that depict illegal or disallowed scenarios. Even the attempt to produce such content is a violation of policy. For instance, prompting an AI with a detailed rape scenario is against our rules – regardless of whether the AI ultimately produces an output or not.

Monitoring and Filters: Fantra employs automated content moderation filters that screen prompts and outputs for policy violations. If you attempt to upload disallowed content, you will likely receive a warning or block message from the system. These filters are continuously updated for accuracy. Additionally, Fantra reserves the right to review and audit user-uploaded content for compliance. While we respect user privacy, any content flagged by our systems or reported by users may be reviewed by our moderation team. As noted in our Community Guidelines, we use a multi-moderator “reasonable person” standard for evaluating AI content – particularly in edge cases (e.g., determining if a character looks underage or if a deepfake is of a real person).

No Expectation of Anonymity in Prompts: You should understand that prompt data may be retained and examined for the purposes of improving our filters, preventing abuse, and complying with legal obligations. Do not input any personal or sensitive information in prompts that you wouldn’t want reviewed. All use of the AI must comply with our Privacy Policy and these Guidelines.

Consequences of Attempted Evasion: As stated, trying to circumvent the AI’s safety features is a violation. For example, using euphemisms or slang to generate banned content, or splitting a prompt into pieces to slip past filters, will be detected and may result in disciplinary action. The system may issue warnings or require acknowledgment of the rules if it detects minor infractions. Repeated or serious attempts to violate content rules via clever prompting will lead to account suspension. We treat circumvention attempts the same as actual content violations – both are breaches of trust.

User Responsibility: You are responsible for the content your prompt the AI to create. Even if the AI system produces an unexpected or disallowed result, you should not share that result and should report it if it clearly violates policy. Likewise, if you notice a loophole in the AI that allows forbidden content, report it rather than exploiting it. Remember, “I didn’t intend this” is not an excuse if you then knowingly save or distribute the disallowed content.Always use the AI tools in good faith according to the rules.

Human Oversight: While AI may automate part of our moderation, Fantra’s Trust & Safety team provides human oversight. If your content is flagged and you believe it’s a mistake, you can reach out as described in the Appeals section. Conversely, moderators might remove content that the AI initially allowed if upon review it’s found non-compliant. The AI is not perfect – the final judgment lies with our human moderators and these Guidelines.

Model Updates and Future-Proofing

AI technology and norms are rapidly evolving. Fantra is committed to continuously updating our AI models, filters, and policies to address new challenges. These Guidelines are intended to be future-proof, meaning they apply to current AI features and any new AI capabilities we introduce:

• Model Improvements: Fantra will regularly update its AI models to improve quality and enforceability of these Guidelines. This may include training the AI to better refuse disallowed prompts, improving image recognition of prohibited content (like detecting if an image looks underage), and refining prompt processing. We will also adjust what content the AI can generate as needed for safety. For example, if new forms of AI content emerge (e.g. AI-generated video), the same rules will apply to them. We may explicitly expand the policy to cover new media types as they arise.

• Adaptation to Laws: We track legal developments worldwide concerning AI and content. If laws change or stricter regulations apply (for instance, new laws against deepfakes or AI-generated explicit material), Fantra will update these Guidelines and our practices to ensure compliance. We include broad prohibitions now (e.g. against non-consensual deepfakes) to align with the spirit of emerging laws, but we may further tighten rules as needed. Users will be informed of major changes, but it is your responsibility to stay updated with the latest version of our policies.

• User Education: We aim to provide resources and updates to help users understand what is allowed. Future iterations of Fantra may include in-app tips or required tutorials for AI content creation to reinforce these Guidelines. Safe use of AI is a shared goal – we appreciate the community’s cooperation in identifying issues and suggesting improvements.

• No Grandfathering of Old Content: If we update the Guidelines or our AI systems such that something previously allowed becomes disallowed, we will typically require its removal even if it was created before the change. Fantra reserves the right to remove or re-classify older content to meet new standards. We will try to give notice or time for creators to adjust, but our priority is platform safety and compliance. In short, compliance is an ongoing requirement – staying within the rules is not a one-time thing but a continuous obligation as standards evolve.

By using Fantra, you acknowledge that AI technology and rules may change and agree to use the tools in a manner that remains compliant with the latest Guidelines. We encourage you to review these Guidelines periodically. The “Last Updated” date at the top indicates when changes were last made. If you’re ever unsure about new features or content, check the official Fantra Legal/Policy page or contact support.

Enforcement and Appeals

Violations of these AI Content Guidelines are taken seriously. Fantra enforces these rules to maintain a safe and legal platform. Enforcement actions can include any of the following, depending on severity and history of the violation:

• Content Blocking or Removal: Our AI filters may automatically block prohibited content from being created. If disallowed content is somehow generated or posted, we will remove or disable access to it without prior notice. In some cases, we may warn you and allow you to correct the issue if it’s borderline, but we are not obligated to do so. Strictly forbidden content (e.g. anything sexual involving minors) will be removed immediately and typically without warning, and likely reported to authorities.

• Warnings and Temporary Suspension: For a first-time or less severe violation, Fantra may issue a warning to you. You might be required to acknowledge the rules again and promise compliance. We may also impose a temporary suspension (“cool down period”) during which you cannot generate or post content. This is to give you time to understand the breach and prevent further issues. Multiple minor violations can accumulate to more serious action.

• Account Restriction or Termination: Serious violations (for example, attempting to create CSAM, repeated harassment via AI, or creating non-consensual deepfakes) or repeat offenses will result in your account being suspended or permanently terminated. If your account is terminated for violating these Guidelines, you are barred from using Fantra again. We may implement technical measures (such as banning associated emails, device or payment identifiers) to prevent re-registration. Account termination for cause may also result in forfeiture of any unused subscription or virtual currency balance, in accordance with our Terms of Service and applicable law .

• Reporting to Law Enforcement: **Any content or use of Fantra that involves potential criminal conduct – especially exploitation of minors (even attempted) or other egregious illegal content – will be reported to the appropriate authorities. Fantra will cooperate with law enforcement investigations. We also comply with legal orders such as subpoenas or warrants if required to disclose user information related to prohibited content. Don’t risk it – the consequences of generating illegal content go beyond just losing your account.

• Legal Action: Fantra reserves the right to pursue legal action against users who violate these Guidelines in ways that cause harm to others or to Fantra. This could include civil lawsuits for damages or injunctions. For example, creating and disseminating a fake explicit image of someone could lead to legal liability for defamation or privacy breaches, and Fantra may assist victims in such actions. While we hope not to involve courts, we will use all remedies available to us to enforce critical rules.

Appeals: If you believe that a content removal or enforcement action was made in error, you have the right to appeal. You can contact Fantra support or follow the process in our Complaints Policy to request a review. Please provide any relevant information (e.g. why you think your content was compliant, or context that moderators might have missed). Our team will re-evaluate the decision. Minor violations may be given a second look and possibly reversed if justified. However, major violations (such as confirmed child exploitation content or clear non-consensual porn) are not eligible for appeal – those decisions are final. In all cases, Fantra’s determination on appeal is at our sole discretion and constitutes the final word.

During an appeal review, your account may remain suspended or limited as a precaution. We strive to handle appeals promptly and fairly. If a mistake is found, we will restore content or access, and it will be as if the action never happened (no strike on your record). If the action stands, you will be notified, and the enforcement will continue.

Reporting Issues: We also encourage users to report any content or behaviour that seems to violate these AI Content Guidelines or other policies. If you see something concerning, use the in-platform report function or email us. We rely on both automated systems and community vigilance to keep Fantra safe. Good-faith reports are appreciated and will not result in any penalty to the reporter.

By following these AI Content Guidelines, you help keep Fantra a safe, legal, and creative space for AI-generated expression. We thank you for your cooperation and for using AI responsibly. If you have any questions about whether something is allowed, please reach out to admin@fantra.ai before proceeding – we’re here to help clarify. Happy creating, and remember to always keep it consensual, adult, and within the rules!