Acceptable Use Policy

1. Scope

This policy is part of the Flowwyn Terms of Service. It covers messages, imports, files, generated replies, and content stored or shared through the service, including private chats. Listing a content format does not mean that generation feature is currently offered.

The beta is US-only, by invitation, and for adults 18 or older. Adult eligibility does not permit prohibited sexual content.

2. Prohibited content categories

Do not request, create, upload, store, or distribute content in any of the following six categories through Flowwyn:

1. Pornography or NSFW content: sexually explicit material or sexual content generation, including sexual exploitation and non-consensual intimate imagery, whether authentic, altered, or generated. Do not create or threaten to distribute intimate content without consent.
2. Violence or gore: graphic depictions of serious injury, torture, sexual violence, or cruelty for shock or gratification; credible threats; encouragement of real-world violence; instructions facilitating attacks; or encouragement of suicide or self-harm.
3. Hate speech: promotion of hatred, dehumanization, discriminatory abuse, or violence based on characteristics such as race, ethnicity, nationality, religion, sex, gender identity, sexual orientation, disability, or caste.
4. Child-unsafe content: child sexual abuse material (CSAM), whether real, generated, fictional, or role-played; any sexualization of anyone under 18; grooming or exploitation of minors; content that encourages minors to harm themselves or others or to take part in dangerous or illegal activity; and any assistance facilitating these acts.
5. Deepfakes or impersonation: fabricated appearances, voices, statements, or actions of real people through face swaps, voice cloning, nudification, or similar techniques, even if the person consents. Do not impersonate people or organizations deceptively or falsely imply endorsement.
6. Copyright or trademark infringement: unauthorized use of protected work or branding, bypassing access restrictions, or passing off another's work as yours. Flowwyn does not offer features for generating protected characters or reproducing real people's likenesses.

Do not upload or share photos, videos, or voice recordings of another person without that person’s consent, and do not use Flowwyn to edit, imitate, or swap someone’s face or voice. Flowwyn does not offer face-swap, voice-cloning, or likeness-generation features.

Lawful, neutral, non-graphic education, prevention, reporting, or help-seeking discussion is not automatically prohibited. Context does not excuse CSAM, sexual exploitation of minors, non-consensual intimate imagery, prohibited deepfakes, or instructions facilitating serious harm. Provider rules may be stricter.

3. Other prohibited uses

• Violent extremism or terrorism: recruitment, propaganda supporting violence, operational assistance, or material support.
• Weapons of mass harm: designing, acquiring, producing, or using chemical, biological, radiological, nuclear, or other means of mass violence.
• Harassment: stalking, doxxing, blackmail, coordinated abuse, targeted intimidation, or threats.
• Malware and cyberattacks: malicious code, credential theft, account compromise, unauthorized surveillance or exploitation, and service disruption.
• Fraud and spam: scams, phishing, identity theft, payment fraud, forged documents, deceptive mass messaging, fake reviews, or passing off AI output as verified human professional work.
• Illegal activity: unlawful transactions, trafficking, evasion of legal restrictions, or other criminal conduct.
• Privacy violations: using or exposing private, confidential, biometric, or sensitive information without necessary rights or lawful authority; uploading stolen archives or others' credentials.
• Safety circumvention: evading moderation, account/age/location/invitation controls, billing restrictions, or provider safeguards, including repeated reformulations of blocked harmful requests.
• Reselling access: selling or sharing accounts, invitations, credentials, or access; transferring/reselling credits; or using Flowwyn to top up or sell access to third-party AI accounts.

Do not use AI alone for consequential decisions about another person's health, credit, employment, housing, legal rights, or similar interests without lawful safeguards and appropriate human review.

4. Automated checks and enforcement

The beta will check text before it is sent to an AI model and check each reply before it is shown. Before sending, the text is checked by OpenAI’s Moderation API, by Waffo’s Prompt Sift screening service, and against Flowwyn’s rules, which we update regularly. A request is sent only if all of these checks allow it. Each AI reply is checked in full by Flowwyn’s own moderation, using OpenAI’s Moderation API and Flowwyn’s rules, before it is displayed. Paid AI access will remain unavailable until the checks are enabled and verified.

The checks cover new submitted text and AI-bound context not already checked, such as imported messages, extracted file text, project notes, and About you. Stored archives, drafts, and profile edits are not scanned on their own. This is not comprehensive scanning of uploads or storage and is not a dedicated CSAM detection system. Known or suspected CSAM must not be sent to an AI provider. We keep records of check results and of human review decisions, without the checked text, as described in the Privacy Policy.

If a check blocks a request, it is not sent. If a required check cannot give a result, the request is not sent and your draft stays so you can try again later; if a check on a reply cannot be completed, the reply is withheld.

When the checks identify signs of self-harm intent in a message, Flowwyn shows a referral to call or text 988 in the United States and to call 911 if there is immediate danger. Checks can miss relevant content. Flowwyn is not an emergency service, and support is not monitored around the clock.

Checks can miss violations or flag permitted content. A flag is not proof of wrongdoing; passing a check does not make a use permitted. Authorized reviewers may review reports and available records and context, record decisions, and take proportionate action. We may restrict requests or suspend access. For a confirmed reported reply, we can remove Flowwyn’s improvement copies, prevent future AI use of that reply, and invalidate existing export downloads. The original reply can remain visible in your own conversation. We may also make legally required reports. Copyright notices follow the process in the Terms.

Inputs blocked before reply generation do not incur a reply charge. A generated reply that is withheld by safety checks is still charged for the AI usage it took, including when a safety check cannot be completed. Choose Report and “This reply shouldn't have been blocked” to request review. If support confirms an incorrect block or a safety-check failure, we credit that reply’s charge back to your Flowwyn balance.

The Privacy Policy explains safety processing and disclosures. Turning off sharing for improvement does not disable safety checks or report handling.

5. Provider policies

Applicable provider restrictions also apply. OpenAI, Anthropic, and DeepSeek, only when enabled with your opt-in, may apply their own safeguards. Switching models does not create permission for prohibited use.

Provider references: OpenAI Usage Policies · Anthropic Usage Policy · DeepSeek Open Platform Terms and linked policies.

Providers may independently block requests, retain abuse records, suspend access, or report suspected unlawful activity. Flowwyn does not control those decisions and remains responsible for its own duties.

6. Report a concern

When the beta opens, you will be able to choose Report on a reply generated in Flowwyn. You can also email support@flowwyn.com with the subject “Report.” To report an imported reply, email support with its conversation reference. Flowwyn does not retain withheld reply text; support reviews the available records and context.

Include a conversation/reply reference, the reason, and enough information to locate the issue. For privacy or intellectual-property complaints, identify the affected right or work and your relationship to it. Do not attach or forward suspected CSAM or other illegal material; describe where it appeared instead. Do not send passwords or sign-in codes.

7. Reports, decisions, and appeals

During the beta, Flowwyn’s founder will handle reports received through support@flowwyn.com. Reports are triaged with child safety and credible threats first. We review relevant context and decide whether to restrict content or access or take other action. We record decisions and may ask for clarification. Where appropriate and lawful, we notify reporters or affected users of the outcome.

Access to report information is limited, but confidentiality cannot be absolute: investigation, legal process, or required reporting can involve disclosure. When the reporting conditions in 18 U.S.C. § 2258A are met, we report apparent child sexual abuse material, child sex trafficking, or online enticement of a minor to NCMEC as soon as reasonably possible and preserve the reported material and required context for one year after the report. We do not treat every safety flag as a crime.

To appeal, email support@flowwyn.com with the subject “Appeal”, the reference, and your explanation. We will review and correct mistaken decisions where warranted. Flowwyn is not an emergency service; reports and appeals are not an emergency channel. Contact emergency services for immediate danger.

8. Changes and contact

We notify account holders by email and in-app notice at least 14 days before material changes take effect, following the notice process in the Terms.

Flowwyn 444 Alaska Avenue, Suite #CHT477 Torrance, CA 90503, USA support@flowwyn.com

Related pages: Terms · Privacy · Pricing · Consumer Health Data