What we know. What we are still testing.

What you type here is your own check-in. Daylogue's AI reads it, like anyone's. Case notes belong in the caseload.

Describes current processing only. Says nothing about retention or who in the department can see the check-in.

Psychological Services can include you in its follow-up after a critical incident you were part of. Turn it off and you are left out.

Describes the follow-up survey audience. Does not cover peer outreach, which has its own switch.

Public Group membership and school enrollment are separate; school completion uses explicitly submitted work and teacher confirmation.

Only the described implemented product restrictions are supported. Live school activation, legal compliance, research authorization, wellbeing measurement and native distribution are not asserted.

Groups shares only deliberately submitted episode responses with the organizer; group membership does not expose private journals or AI conversations.

Only the described pilot behavior is substantiated. No privacy guarantee, legal-compliance certification, wellbeing measurement, research authorization, or teen AI activation claim is supported.

Teen enrollment does not itself activate AI or give another person access to account content.

Corrects the obsolete blanket guardian requirement and narrows sharing access. Does not announce a live cohort or certify legal compliance or safety. Expanded/promotional claims remain held.

Daylogue reads the entries you choose to sync to create narratives and surface patterns. Entries are not end-to-end encrypted.

This is a corrective limitation, not a claim that Daylogue is more secure than another product.

Data is protected in transit with TLS, and ordinary account access is restricted with per-user database controls.

Describe controls, never an impossible security outcome.

Daylogue does not use personal entries to train its own models.

Must remain scoped to Daylogue model training.

Daylogue does not infer emotion from faces, voice tone, or physiology. It works from the words and context people choose to share.

The claim must be re-reviewed whenever a new sensor, voice, camera, or health feature is introduced.

Account deletion begins with a 30-day grace period when you can cancel the request. After that period, Daylogue begins deletion from active product systems. Backup, processor, billing, audit, legal, and user-authorized Collab retention limits are described in the Privacy Policy.

This wording describes a phased lifecycle and does not promise deletion from every layer on one fixed day.

When you choose to connect Apple Health, Daylogue imports eligible records as context. Daylogue does not measure sleep, calculate a readiness score, or predict performance.

Only eligible records verified on shipped Apple surfaces may be named.

Employer-context dashboards are designed not to display individual entry text, transcripts, member-level mood, energy, or stress scores, or individual participation histories. Eligible aggregate self-reported trends appear only after the applicable contributor and suppression threshold is met.

Do not use this record to describe non-employer roles or consent-based Collab sharing.

Daylogue does not sell personal data. Disclosed service providers process data to operate requested features.

Corrective operational statement only. It does not decide whether a particular disclosure constitutes sharing, targeted advertising, or another regulated transfer in a jurisdiction.

Before a covered conversational check-in begins, Daylogue identifies the feature as AI and states that the user is interacting with an AI system, not a person.

Operational transparency statement only; not a conclusion that a named statute applies or has been satisfied.

Some information Daylogue processes may be treated as consumer health data under state law depending on the information, how it is used, the person's location, and whether a law applies. Daylogue provides a separate Consumer Health Data Privacy Policy and rights process for transparency. This does not claim that every record has the same legal classification in every jurisdiction.

Corrective uncertainty disclosure and voluntary rights-process scope only.

Daylogue welcomes good-faith vulnerability reports. We review reports and coordinate next steps based on severity, impact, and available information. This page does not grant authorization to test third-party systems, create a fixed disclosure deadline, promise immunity from legal action, or permit activity that violates applicable law.

Good-faith reporting invitation only. No legal immunity, authorization, fixed SLA, payment, or disclosure deadline.

Daylogue keeps its IPIP-50 means, Reflection Profile, MBTI, and DISC as separate frameworks and does not convert one result into another.

Approved scope covers framework separation only, not psychometric validation, professional-care claims, or equivalence between instruments.

Separately authorized Daylogue support, security, incident-response, and legal-process access follows the scoped controls and disclosures described in this policy; a person may also choose a separate sharing path where the product offers one.

Corrective limitation only. It does not create or expand an access permission.

Daylogue is a system for self-understanding, not therapy, a coaching service, or an HR tool, and it is not a replacement for qualified professional care or people-development work.

Only scope and referral wording is approved. No assessment, diagnosis, treatment, coaching, mediation, or employment conclusion is implied.

Daylogue shows private support resources without automatically notifying an employer, school, organization, trusted contact, crisis service, law enforcement, or emergency service, and without submitting a workplace or school concern on the person’s behalf.

Corrective operational disclosure only. It does not create an emergency-response service or an absolute promise against legal process or separately authorized access.

GTM and GA4 are mounted only on public funnel layouts, and Global Privacy Control or the Daylogue Do Not Sell or Share control suppresses GTM and marketing attribution.

Corrective data-flow disclosure only; it is not a universal no-third-party-processing claim.

A Big Five score must be interpreted with the named instrument, scoring key, and reference sample; support for the broad model does not validate every online quiz that uses the OCEAN labels.

Educational interpretation boundary only; no instrument or quiz receives a blanket validity endorsement.

Turning the self-model off deletes what Daylogue noticed — the observations, the scores derived from them, and the answers given to them. The check-ins those observations were read from are not deleted, and the record that consent was given and withdrawn is kept.

The wording promises deletion of the derived self-model records only. It is not a claim about backups, processor copies, the source check-ins, or the append-only consent audit.

Daylogue can report that a recent window of someone's own writing differs from their own earlier baseline. It is a difference between two periods of the same person, carries no direction, and is not a measurement of health, mood, or capability.

Within-person temporal difference only. No directional, comparative, or clinical reading is approved.

The original text is stored only until you confirm or cancel the import, then purged.

Describes the shipped review-then-purge lifecycle. It is not a promise of instantaneous deletion from every storage layer, and it does not cover text already sent to an AI provider for extraction.

No AI mode is on. Your imports are saved and dated, and theme reading stays off until you turn AI back on.

Describes the shipped behavior of the Backstory hub while No AI mode is on. OCR scanning is unavailable in that mode.

Your current privacy settings require writing in your journal on the web.

Narrow disclosure of SMS incompatibility with existing storage choices. No new encryption guarantee; live handset verification remains a canary requirement.

Pro includes more guided journaling and eligible AI reads of writing you choose to share.

Describes existing plan capabilities, not new prices, unlimited AI access or demonstrated retention. Live SMS correctness and later conversion observation remain separate.

Your coach sees what's affecting the team, never what you wrote.

Describes coach-role access in sports programs. Programs created before 2026-09-26 may still have individual sharing on until reviewed; the promise holds there because sharing still needs each athlete's own consent. Does not describe Collab, employer admin roles, or authorized service, support, security, and legal-process paths.

Program license, for athletic departments and schools. Custom quote.

Describes availability and packaging, not a price, contract term, or outcome. The education FAQ pilot sentence covers team pilots generally and rests on the same pilot-length table. Provisioning at the enterprise tier also carries Enterprise-only controls; the license is not marketed on them.