befall United States Privacy Notice
This notice at collection explains the personal-information categories, purposes, training boundaries, retention criteria, and state privacy rights associated with befall. befall identifies PROJECT82 LLC as its intended operator. That entity remains formation pending, so this draft cannot become effective until formation and the contracting-party record are verified. Its verified business mailing address is 11F, 16, Teheran-ro 20-gil, Gangnam-gu, Seoul, Republic of Korea; no unverified Wyoming address is presented as a registered office.
Who is responsible for the data
befall identifies PROJECT82 LLC as its intended operator. That entity remains formation pending, so this draft cannot become effective until formation and the contracting-party record are verified. Its verified business mailing address is 11F, 16, Teheran-ro 20-gil, Gangnam-gu, Seoul, Republic of Korea; no unverified Wyoming address is presented as a registered office. The identified operator determines why and how covered personal information is processed for this service, subject to the draft and release limitations stated here. This is a release-blocked draft with no approved effective date. Before publication, PROJECT82 LLC formation and its Wyoming registered-office record must be verified, a DMCA designated agent and working legal/DMCA mailbox must be confirmed, vendor facts must be reconciled to live configuration and contracts, and United States counsel must approve the state-law, arbitration, consumer, and sales disclosures.
Notice at collection: categories collected
befall provides interactive episodic fiction shaped by typed lines and choices. Story progress, selected choices, typed lines, and safety signals support continuity and moderation. befall sells the Story Pass monthly subscription at US$9.99, which grants 1,200 coins for each paid billing period; a saved answer or a reroll costs 8 coins, and published launch episodes can be read without payment. Entry begins at age 16; ages 16–17 use the safety tier, and training eligibility begins at age 18. The adult tier begins at age 18, and data from younger or unknown-age users is excluded from training. Depending on use, categories may include account and contact details; age or birth-date data; device, log, and cookie data; conversation or call content; recordings and derived transcripts; character, memory, and preference data; safety reports; purchase and transaction records excluding full payment credentials handled by a store or payment provider; and support communications. A category not verified in the product data flow is not added to this notice by assumption.
Sources of personal information
We receive information directly from you, from your interactions with AI and safety features, from the device or browser you use, from app stores or payment services, and from service providers acting for the operator. We do not treat an unverified vendor inventory as a final public recipient list.
Purposes and authority for use
We use information to create an account, provide requested AI responses and continuity, classify age, personalize features, process purchases, detect abuse, investigate reports, secure systems, provide support, comply with law, establish or defend claims, and, for eligible adult data after valid notice, evaluate and improve models. Applicable consent, contract, legal-obligation, and legitimate-purpose requirements are honored by jurisdiction.
Service processing is separate from training
Operational processing may continue after a training opt-out when needed to generate replies, store conversations or requested memory, perform age and content-safety checks, handle reports, prevent fraud, secure accounts, complete payments, provide support, or meet legal duties. Training, fine-tuning, evaluation-dataset addition, and long-term pseudonymized retention for those purposes stop prospectively as described in the AI Data Use Notice.
Human review and labeling
Only a limited number of authorized personnel may review the minimum context needed for improvement quality evaluation or labeling. Improvement labeling views must never expose account or direct contact details. Operational support or safety review is a distinct limited-purpose process and does not make those details available for improvement labeling. Access, viewing, export, and deletion events are logged, and an external labeler may not be used until confidentiality, security, access, and deletion obligations are contractually verified.
Service providers and disclosures
Service providers may support hosting, AI inference, security, customer support, analytics, communications, or payments only under appropriate instructions and safeguards. The provider names, locations, purposes, transfer details, and retention terms are not yet verified against live configuration and contracts; this notice remains blocked until that roster is complete. We may also disclose information when lawfully required or necessary to protect rights and safety.
Cross-border processing
Information may need to be processed outside the state or country where a user lives. Before this notice becomes effective, the operator will verify each destination, recipient role, purpose, transfer mechanism, safeguard, and retention term. This draft does not substitute an assumed country or vendor for that required transfer record.
Retention criteria
Retention criteria include the time needed to provide an active account or requested feature, the sensitivity and volume of the data, safety and fraud needs, legal recordkeeping, dispute limitation periods, a valid preservation request, and the time required for secure deletion from backups. Training copies follow the additional eligibility, opt-out, and separability rules in the AI Data Use Notice rather than an unlimited retention period.
Security safeguards
We use administrative, technical, and organizational measures proportionate to the data and risk, such as access restriction, logging, separation of operational and training stores, masking, and protected transmission where appropriate. No system is perfectly secure, so this notice does not promise absolute prevention of loss or unauthorized access.
State privacy rights and choices
Depending on residence and applicable law, you may request access, correction, deletion, or portability; opt out of certain sale, share, or targeted advertising activity; limit certain sensitive-data use; and receive equal service without unlawful retaliation. You may appeal a denied request and may use an authorized agent where permitted. This draft does not authorize the monetary sale of training data. A factual classification of any SDK disclosure as a sale, share, or targeted advertising event requires completion of the vendor audit before release.
Children, teens, and unknown ages
Entry begins at age 16; ages 16–17 use the safety tier, and training eligibility begins at age 18. The adult tier begins at age 18, and data from younger or unknown-age users is excluded from training. All minor and unknown-age events are ineligible for model training, evaluation datasets, and labeling for improvement. If age information changes or indicates a minor, pending training material is removed or blocked under the AI Data Use Notice.
AI outputs and automated processing
Automated systems generate content, personalize continuity, classify safety risk, and help detect abuse. They may make mistakes and are not represented as human judgment. Where applicable law grants review or opt-out rights for a decision producing legal or similarly significant effects, the operator will provide the required process rather than relying on this general description.
How to make a privacy request
A verified privacy and legal request channel will be displayed before this notice becomes effective. The final mailbox has not yet been confirmed, so this draft does not invent contact details. Requests may require proportionate identity verification, and an authorized agent may need to show authority.
Changes to this notice
We will date and describe material changes and provide prominent advance notice when required. New training terms apply only after the required versioned notice and never backdate authority. Earlier versions remain in revision history, and mandatory rights continue despite a policy change.
Document history
- 2026-07-27-draft.1 · 2026-07-27 — Initial PROJECT82 regional legal draft.
Last updated 2026-07-27 · content f6f078967fae · 대한민국 · 한국어 · 日本 · 日本語