Age-appropriate use
Swerve is not designed for young children. Users under the age of majority should use it only with parent or guardian consent.
Swerve safety
Swerve is an AI character chat and roleplay experience. These guidelines explain the basic boundaries we use to protect users, support reporting, and keep the product safer.
Swerve is not designed for young children. Users under the age of majority should use it only with parent or guardian consent.
Characters may feel personal, but they are AI-generated. They are not real people, therapists, doctors, lawyers, or emergency services.
If you see content that appears illegal, abusive, exploitative, or otherwise unsafe, contact support so the team can review it.
Avoid sharing passwords, financial information, government IDs, addresses, or other sensitive personal data in chats.
Content boundaries
Crisis support
If you or someone else may be in immediate danger, contact local emergency services now. If you are in the United States or Canada, call or text 988. If you are in the United Kingdom or Ireland, you can contact Samaritans at 116 123. For other countries, visit Find a Helpline to locate local crisis resources.
Find a HelplineFor safety reports, account questions, privacy requests, or data deletion help, email [email protected] with enough detail for the team to review the request.
Contact supportEffective date: August 3, 2026
Team: Swerve Team
Service: Swerve — https://swerveai.app
Safety contact: [email protected]
Swerve Team ("we," "us," or "our") operates Swerve, an AI character chat and roleplay service. Some features may support mature content only for users who have completed the applicable adult-access controls. We permit lawful adult expression, but prohibit illegal, exploitative, non-consensual, deceptive, or unsafe content. This Policy applies to prompts, uploads, AI outputs, public and private content, profiles, character descriptions, tags, search terms, messages, advertisements, and conduct connected to the Service.
For this Policy, a Content Provider is any person who uploads, publishes, sells, supplies, or directs the creation of content. A Depicted Person is any identifiable real person whose image, voice, likeness, performance, or personal data appears in content. Public Content includes content accessible to another user, whether free, paid, searchable, shared by link, or downloadable.
Mature features are available only to persons aged 18 or older and, where local law requires, only to persons who have reached the higher applicable age of majority. We use age-assurance controls appropriate to the user's location and the risk of the feature. We may deny or suspend mature-feature access where adult status cannot be established.
Access to mature features is blocked until the applicable age-control step is completed. We may require renewed confirmation following account recovery, material account changes, suspicious activity, a legal requirement, or expiration of verification evidence. Users may not lend, transfer, or share an adult-enabled account with an underage person.
Sexual content may never depict, describe, role-play, sexualize, or appear to involve a minor. This includes fictional, animated, AI-generated, age-regressed, "barely legal," school-age, child-like, or age-ambiguous persons. If age is uncertain, the content is treated as prohibited.
Users and content providers must not request, create, upload, publish, market, sell, share, or facilitate:
Prohibited content remains prohibited when it is synthetic, fictional, private, encrypted, altered, or framed as role-play, parody, fantasy, or education.
Swerve does not currently operate live streaming or a verified adult-performer marketplace. Users may not use Swerve to publish sexual content depicting an identifiable real person unless they hold all required rights and valid consent. We may require proof of identity, age, rights ownership, and consent before allowing or restoring such content.
If Swerve later allows third-party providers to publish, sell, or stream adult content, publication will remain disabled until the provider and every depicted person have completed appropriate identity, age, rights, and consent verification. Required records may include valid government-issued identification, document-holder checks, specific written consent, content identifiers, permitted uses, and withdrawal or dispute procedures.
Blanket, implied, coerced, intoxicated, or unverifiable consent is not accepted. Failure to provide requested supporting records results in immediate restriction of the affected content and may result in account termination. Verification and consent records are retained securely only for as long as required by law, contract, investigation needs, payment-partner requirements, or our documented retention schedule, and access is limited to authorized personnel.
AI prompts, reference uploads, metadata, search terms, character context, and generated outputs may be screened using layered automated controls and risk-based review. High-risk, uncertain, or flagged content may be blocked, quarantined, restricted, or escalated for trained human review. A blocked item may not be published through manual workarounds or alternate accounts.
Review considers the content and its surrounding context, including titles, tags, character age, marketing copy, account history, and attempts to evade controls. Automated age estimation alone is not accepted as proof that a depicted person is an adult.
We do not market the Service, permit metadata, or intentionally return search results that imply child exploitation, age ambiguity, trafficking, incest, or non-consensual activity.
Where a feature requires pre-publication review by law, contract, or payment-network rules, publication remains disabled until that review is complete. Material edits, replacement media, newly added participants, or changed metadata may trigger a new review before republication.
We monitor public and generated content through user reports, automated detection, risk-based sampling, trend analysis, repeat-offender signals, and human review. We may restrict visibility, pause generation, preserve evidence, or remove content while a review is pending.
Post-publication controls may include repeat scanning, complaint-triggered review, sampling of approved content, monitoring of prohibited keywords and evasion patterns, and review of accounts associated with prior violations. Confirmed violations may be used to improve detection rules, moderator guidance, and risk scoring.
The Service does not currently provide live streaming. Before any live-streaming feature is launched, it must support verified-adult participation, real-time monitoring, immediate stream termination, evidence preservation, and prevention of immediate restart after a serious violation.
Anyone may report suspected illegal content, policy violations, impersonation, non-consensual content, trafficking, or rights infringement by emailing [email protected]. A report should include the content or character location, reason for concern, relevant dates, and available supporting evidence. Anonymous reports are accepted where technically and legally feasible.
Each report receives a case record and is triaged according to severity. We aim to acknowledge reports within one business day and resolve all content complaints within seven business days. If a final determination requires more time because of law-enforcement involvement or exceptional complexity, we record the reason, keep appropriate safeguards in place, and provide a status update where legally permitted.
Content that may be illegal, depict or exploit a minor, be non-consensual, involve trafficking, or present an imminent risk is disabled or otherwise made unavailable as soon as reasonably practicable pending review. Confirmed illegal or prohibited content is removed without delay. We do not require a reporter to obtain a court order before we take protective action under this Policy.
Complaint records include the intake channel, content identifier, allegation, severity, actions taken, decision, timestamps, reviewer, notifications, appeal status, and closure date. Reporters and affected persons are protected from retaliation, and good-faith reports do not result in account penalties.
Where we identify suspected child sexual abuse material, trafficking, or other serious criminal activity, we preserve relevant evidence and report to the legally designated hotline, law-enforcement body, or competent authority as required. We cooperate with valid legal requests and payment-partner investigations.
We do not download, duplicate, or circulate suspected illegal material except where necessary and legally authorized for evidence preservation and reporting. Access is restricted to designated personnel, and evidence handling is logged.
Any person who appears or is reasonably identifiable in content may request removal by emailing [email protected] with the subject line Depicted Person Removal Request. We may request proportionate information to confirm identity or authority.
After a request, we assess whether valid consent exists for depiction, publication, and download. If consent cannot be demonstrated, was exceeded, or is void under applicable law, the content is removed immediately. Disputed content may remain disabled while the review is pending. Retaliation against a person requesting removal is prohibited.
A request may also challenge use of a person's synthetic likeness or voice. We assess the content, source material, authorization, and risk of confusion or harm. When identity or consent remains uncertain, the content stays unavailable. Removal from public access does not prevent limited retention where legally required for evidence, fraud prevention, dispute resolution, or regulatory compliance.
Depending on severity, context, recurrence, and legal duties, we may:
Severe violations—including content involving minors, non-consensual sexual content, trafficking, or deliberate safeguard evasion—may result in immediate permanent termination without prior notice.
Affected users may appeal by emailing [email protected] with the subject line Safety Decision Appeal. Appeals are reviewed by a person not responsible for the original decision where practicable. Reinstatement is not guaranteed.
We use a graduated enforcement model for lower-risk violations, but no prior warning is required. Repeat violations, coordinated evasion, falsified verification records, or links between terminated accounts are aggravating factors. Restored content must pass the same publication controls as new content.
Moderators receive role-appropriate guidance on age ambiguity, consent, exploitation, trafficking indicators, synthetic media, evidence handling, privacy, escalation, and personal well-being. Access to sensitive material is restricted and logged. We conduct quality reviews, update rules as abuse patterns change, and maintain written escalation procedures for high-risk cases.
Swerve Team's Trust and Safety lead owns this Policy. Reviewers have authority to restrict content immediately and a defined escalation route for suspected minors, trafficking, non-consensual content, threats, and law-enforcement requests. Quality assurance may include periodic decision sampling, calibration, error tracking, and corrective guidance.
Personnel may not review a complaint or appeal where they have a conflict of interest. Sensitive review work is subject to least-privilege access and confidentiality obligations.
We maintain auditable records appropriate to the features we operate, including moderation decisions, complaints, removals, appeals, enforcement actions, and required reports. We provide compliance evidence to authorized payment partners or authorities when legally permitted and contractually required.
Where required by a payment provider, we submit requested content-violation reports, including zero-incident reports where applicable, and remediate identified deficiencies. We monitor fraud, chargebacks, transaction velocity, account linkage, sanctions risk, and unauthorized payment use using risk-based controls.
Our compliance register may track:
Records are time-stamped, retrievable, protected against unauthorized alteration, and retained according to documented legal, security, dispute, and contractual needs. We review this Policy and its controls at least annually and after a serious incident, material product change, new adult-content feature, or payment-partner finding.
Safety and verification data is collected only for specified compliance, security, rights-management, and legal purposes. It is protected through access controls, retention limits, vendor safeguards, and appropriate technical and organizational measures. It is not publicly displayed. Our separate Privacy Policy explains data categories, disclosures, retention, and user rights.
Government identification, biometric or liveness data, consent records, and complaint evidence receive heightened protection if collected. We limit collection to what is necessary, document retention periods, maintain deletion procedures, and assess vendors before sharing sensitive data. A privacy request does not require deletion of records that must lawfully be retained for safety, fraud prevention, legal claims, or regulatory duties.
We may restrict features, content, transactions, or access by location to comply with local law, sanctions, payment-network rules, or contractual obligations. Users remain responsible for complying with laws applicable to them; this does not limit our independent duty to enforce this Policy.
We review the availability of mature features by jurisdiction and may apply geoblocking, transaction blocking, feature restrictions, or enhanced age controls. Attempts to evade geographic or sanctions controls may result in termination.
We may update this Policy as laws, payment requirements, technologies, or risks change. Material updates will be posted with a revised effective date.
Report safety concerns: [email protected]
Request content removal: [email protected] (subject: Depicted Person Removal Request)
Submit an appeal: [email protected] (subject: Safety Decision Appeal)
Privacy requests: [email protected]
Legal notices: [email protected]