Legal · US
boyfrnd United States Content and Community Policy
This policy explains permitted creative use, strict safety boundaries, intellectual-property rules, moderation, reports, and appeals for boyfrnd.
Version 2026-07-27-draft.1 · Last updated 2026-07-27 · Not yet in effect (draft)
Allowed creative content
boyfrnd is an adult AI companion service with proactive messages, memory, and relationship stages. Users may create lawful fictional dialogue, roleplay, characters, and stories appropriate to the product's age tier. Mature fictional themes are permitted only within the adult tier, when the product offers one, and remain subject to consent, safety, privacy, and intellectual-property limits.
Sexual content and absolute minor-safety bans
Content that sexualizes, exploits, grooms, traffics, or endangers a minor is prohibited whether the minor is real or fictional. The ban covers abuse material, sexualized roleplay, grooming, and fictional framing, without requiring a connection to real-world abuse. Adult fictional roleplay may be considered for improvement only when it contains no actual user sensitive data and passes safety and rights filters. Minors and users whose adult status is unknown may not use the service or contribute data to training.
Real people, identity, and impersonation
Do not create or distribute deceptive impersonations, non-consensual sexual depictions, false endorsements, or content intended to confuse others about a real person's participation. Clearly distinguish fiction from a real individual, obtain required permission, and do not use identity traits to harass, defraud, or evade enforcement.
Intellectual property and privacy
Submit only material you have the right to use, and do not expose another person's private information. A DMCA notice must include the authorized person's physical or electronic signature; identify the copyrighted work or a representative list; identify the material and its location; provide contact information; state a good-faith belief that the use is unauthorized; and state under penalty of perjury that the information is accurate and the sender is authorized. A counter-notice must include a physical or electronic signature; identify the removed material and its former location; state under penalty of perjury a good-faith belief that removal resulted from mistake or misidentification; provide the sender's name, address, and phone number; and consent to federal district court jurisdiction and service from the claimant. The operator forwards a valid counter-notice to the claimant and may restore material after the statutory waiting period unless notified of court action. The designated agent and submission channel remain unresolved and release-blocking.
Moderation and safety review
Automated filters and trained reviewers may detect, restrict, label, or remove content and may limit a feature or account when reasonably necessary. Review can be imperfect and considers context, severity, age tier, credible risk, law, and prior violations. Emergency-risk handling does not make the service a crisis service.
Reports and appeals
Users and rights holders may report suspected safety, privacy, impersonation, or intellectual-property violations through the verified in-service or legal channel. When appropriate, we provide the affected user a reason and an appeal path. We may preserve necessary evidence while a report, appeal, legal hold, or rights investigation remains open.
Repeated violations and repeat infringer policy
The repeat infringer policy may restrict content or features and terminate access, in appropriate circumstances, for a user who repeatedly infringes copyright. The operator considers valid notices, counter-notices, restorations, court outcomes, and available appeals rather than treating every disputed allegation as a final strike. Severe non-copyright violations may also support proportionate suspension or termination under the other safety rules. The DMCA designated agent and mailbox remain unresolved and release-blocking.
AI-character creation and responsibility
Characters and outputs are generated by AI and may behave inconsistently with a creator's description. Creators must select an accurate age and safety tier, avoid claiming a real person's endorsement, and respect third-party rights. Publication or sharing may be withheld while safety, consent, identity, or ownership concerns are reviewed.
Document history
- 2026-07-27-draft.1 · 2026-07-27 — Initial PROJECT82 regional legal draft.
Operator
PROJECT82 LLC (formation pending)
Business correspondence and legal-notice mailing address: 11F, 16, Teheran-ro 20-gil, Gangnam-gu, Seoul, Republic of Korea
DMCA designated agent: registration pending
jin@project82.kr
Document bundle 2026-07-27-draft.1. Effective dates are pending legal review and LLC formation.