Big Screen Soulmates

Privacy Policy

Last updated: 18 November 2026

Data controller

Birch Bark Incense Inc. (federally incorporated in Canada; registered office 67 Third Avenue, Wawa, Ontario, P0S 1K0) is the data controller for personal information processed through Big Screen Soulmates. Our Privacy Officer is Brian Harbord, reachable at support@bigscreensoulmates.com.

What we collect

  • Account: email address, password (hashed) or OAuth identifier (Google/Apple)
  • Identity: date of birth (used to compute age, enforce 18+, and apply mutual Discover age-range filtering: each member must fit the other's chosen age range to appear for each other — your birth date itself is never shown), display name
  • Profile: gender, who you're seeking, bio, favourite films, relatable characters, actor you resemble
  • Verification photo and screening records: we collect one live-camera pose selfie per session. Two separate models on Amazon Bedrock screen that same photo. Neither identifies you or compares faces. An authorized human moderator may compare a suspicious update with previously approved photos. Apparent age and visual authenticity checks support screening but do not establish identity, age or liveness. Your approved selfie is your only profile photo, visible to signed-in members permitted to see your profile. Records include the assigned pose, capture/session timing, available device-proof status, screening decisions and short reasons, verification status, and a dated record of your consent, its wording/version and device platform. An exact-file SHA-256 fingerprint detects reuse of the identical image file; it cannot reconstruct the photo or match different photos of the same face. We do not request TrueDepth/ARKit face data or intentionally create facial landmarks, depth maps, faceprints or identity embeddings. We do not use face data for sign-in, compatibility scoring, advertising, marketing, sale or face-derived behavioural profiling. Your consent record is included in your data export and deleted with your account.
  • Verification anti-abuse data: the IP address a verification submission comes from, and the number and timing of your submissions. This is used to rate-limit verification and detect multi-account abuse. Your IP address is also recorded in the security log described below.
  • Security log: when you take certain account actions — such as blocking or reporting someone, removing your profile photo, signing in on a new device, changing your password or deleting your account — we record your IP address and device/browser type, to protect accounts and investigate abuse. For a new-device sign-in or a password change, we also email you a security alert showing the device type, approximate country and time. When you delete your account, the record of that keeps a one-way fingerprint of your email address instead of the address itself, so we can later confirm the deletion.
  • Location: your country, city and state/province. During onboarding, the location check records your city and state/province along with your country. We ask your browser for your device location and compare the resulting country with the country of your network (IP) address. A mismatch — the usual signature of a routed or spoofed connection — does not block your account: we show you a banner explaining the conflict and pause your profile from appearing in other members' "same country only" results until the checks agree. Your country is detected automatically and re-checked each time you open your profile page or Discover; you cannot type it in yourself. Location is only read at those moments while the app is open and is not tracked in the background. "Show my city and state/province to other users" starts switched on, so other members see your city and state/province unless you turn it off in your profile. Separately, you can switch on "Use my location for distance" in your profile: this optional setting stores one approximate coordinate (rounded to roughly a 1 km grid, refreshed at most every 6 hours) so you can filter Discover to members within roughly 100 km. That coordinate is used only for that pass/fail check on our servers — no distance is shown to or sent to other members — and switching the option off deletes it.
  • Activity: likes, passes, matches, text messages, unmatches, reports and blocks you make (including reports of individual messages), and your Discover preferences (such as verified-only or shared-favourite settings)
  • Privacy and notification settings: whether incognito mode is on, whether you have match, message and like notifications enabled and whether they're sent by email, to your phone, or both, and which individual conversations you have muted. These are used only to decide who can see you and what we send you — they are never shown to other members.
  • Devices: a device label and the time you last opened the app, so we can show your signed-in devices, enforce the three-device limit, and tell whether an account is still in use
  • Phone notifications: if you allow notifications in the phone app, a notification token for that phone (issued by Apple or Google), linked to your account and that device. It's used only to send you alerts about new matches, messages, likes and Our Movie scenes.
  • Store age range (phone app sign-up): where the law requires it, the app asks Apple or Google for your age range when you sign up. We keep only the answer (for example, 18 or over), which store gave it, how the age was set if the store says, and the date — never a birthday from the store — and use it only to keep under-18s out.
  • Diagnostics: when loading, sign-in or an in-app purchase fails, when you tap "Send error report", or when you submit bug feedback, we collect limited troubleshooting details: the page (without account or conversation identifiers), whether your device was online, device type, software version, app version/build where available, failed step and redacted error text. We group repeated failures in an administrator-only problem inbox and send operational summaries and repeated-failure alerts to the administrator. These reports do not include your private messages, selfies, passwords, payment receipts, precise location or session recordings, and are never used for behavioural tracking. To limit abuse of the public reporting endpoint, a short-lived, hourly changing hash of the network address is used as a rate-limit counter; the original address is not stored in diagnostic reports.
  • Feedback: anything you submit via the Feedback tab

Special-category and sensitive data

Using a dating service may imply information about your sexual orientation. Where GDPR/UK GDPR applies we treat this as special-category data (Article 9) and process it only on the basis of your explicit consent, which you give by creating a profile and choosing who you are seeking. The verification selfie and its hash are treated as sensitive personal information under Quebec Law 25 and, where applicable, as biometric information under laws such as the Illinois Biometric Information Privacy Act (BIPA) and Texas CUBI; we process them for photo screening, human moderation and profile display, retain our copies under the schedule below, and never sell, lease, trade, or share them for profit. You can withdraw consent at any time by deleting your account.

We handle verification selfies with the care that privacy and biometric laws expect: we ask for consent first, limit their use to verification and your profile photo, keep our copies only as long as described here, and remove them through the account-deletion process below. AWS’s exceptional safety handling is described separately. If a law where you live gives you additional rights, contact us and we will honour them.

How we use it

To create your account, verify age and country, show compatible people, enable messaging, generate movie recommendations for you and your matches, moderate the community, and improve the product. We do not sell your personal data, do not "share" it for cross-context behavioural advertising, and do not use it for advertising or automated decision-making that produces legal or similarly significant effects.

Automated photo screening

When you submit a verification selfie, our server sends your current image, assigned pose and screening instructions to two AI image models from different companies, run on Amazon Bedrock (Amazon Web Services). The first screens for content safety, one visible person who appears adult and visible screen or printed-photo indicators. If it passes, the second performs an independent check on the same image, including whether it matches the assigned pose. Neither model receives your earlier photos or the other model’s verdict; neither identifies you, compares faces or searches for duplicate people. We do not include your name, email address, account ID or birth date in the screening content.

Screening models: Google Gemma 3 27B (first check) and Anthropic Claude Haiku 4.5 (second check), both through Amazon Bedrock, which may process them in Canada or the United States depending on the model. We update this line before a model change takes effect.

The first check can pass, reject or flag; the second can pass or flag, and is the only check that looks for AI-generated or edited photos. When either check finds something you can fix, for example sunglasses, poor light, another person or writing in the photo, you are told what to change and can retake for free, up to three times in any 24 hours; after that, the selfie goes to human review. Photos the first check would reject, such as explicit photos or photos that appear to show a child, are not accepted as profile photos. If a check is briefly unavailable because of a usage limit, or fails or gives no answer, it is re-run automatically, and the photo goes to human review if it still can't be checked within 30 minutes. Your selfie stays hidden from other members until both automated checks pass; if either check is unsure, it stays hidden until a moderator decides. A selfie that has passed both checks but is not yet approved is visible with an “Awaiting verification” badge. Automatic approval requires both checks to pass, a verified device proof from the app, and the administrator’s automatic-approval setting to be enabled; otherwise a moderator decides. Moderators can revoke approval. No account is closed by AI alone; contact us to contest a decision.

Our separate 18+ date-of-birth gate supports the adult-only requirement. Visual age and authenticity screening has limitations, so it is supported by server-controlled capture timing, available device proof and human review. These controls assess the submission and capture session, rather than certify identity, age or liveness. Each model’s decision and a short reason are stored with the verification record. Authorized moderators may compare a suspicious update with previously approved photos; this comparison is human-only.

Before capture and upload, we ask for explicit consent covering automated screening, transfer to AWS including processing in Canada or the United States, human review and comparison, profile display, retention and withdrawal. Updated consent is obtained before the replacement process is used. You can take a break before submitting a selfie. A selfie submission is required to use the Discover deck and appear in other members’ decks; an approved badge is not required. Paid conversations normally follow matches formed through Discover. Additional verification submissions after the five free attempts per 12-month period (counted from your first selfie) use a separately purchased consumable credit.

We send your current selfie and screening instructions to AWS Amazon Bedrock. Our selected models may process them in Canada or the United States. AWS states that model providers do not receive customer prompts or completions, and Bedrock inputs and outputs are not used for model training. Authorized moderators review photos; your profile photo is visible to signed-in members permitted to see your profile.

AWS documents that Converse does not store submitted content in ordinary processing. AWS separately may store and review image inputs flagged as potential child sexual abuse material and make legally required reports. AWS does not publish a retention period for those exceptional copies; they are governed by AWS’s safety procedures and applicable contractual and legal obligations. Account deletion removes the copies we control under our application schedule. AWS-held exceptional copies are handled separately under those procedures and obligations. We do not use selfies for model training.

Any model-invocation content logs are additional application-controlled copies with separate deletion or expiry controls. Disabling content logging does not itself remove earlier logs.

AWS’s published Customer Agreement, Service Terms and Data Processing Addendum set out purpose limitations, confidentiality and data-protection obligations. See the Converse reference, AWS safety-handling documentation, Customer Agreement, Service Terms and Data Processing Addendum.

Cookies and local storage

We use a small number of first-party cookies and browser local storage entries strictly to keep you signed in, remember your preferences (e.g. cookie choice), and secure the session. We do not use advertising cookies, third-party analytics trackers, or cross-site tracking. A consent banner is shown on your first visit to the website; the native mobile app stores only the sign-in session and local preferences, so no cookie banner is shown there. You can change or withdraw your choice at any time:

If you arrive from one of our ads, the website keeps that ad campaign's name in your browser for that visit only and adds it to the App Store and Google Play links, so the stores can tell us how many installs each campaign brought. It contains no personal information and is not sent to our servers.

Who we share it with (subprocessors)

  • Other users: your profile, photos, and messages are visible to people you match with or appear to in Discover. If you like someone, your profile (including your photo, unblurred) appears in their Likes tab under "Likes received", even before they like you back. Unmatching hides the conversation but the underlying like record is kept. A finished Our Movie script can be shared outside the app only after both co-writers tap "OK to share"; the shared text contains the film's title and scenes with a link to Big Screen Soulmates, never names or photos, and either writer can turn sharing off again (copies already shared outside the app can't be recalled).
  • Sample profiles: the app contains a small number of clearly labelled sample/test profiles (marked with a "Sample profile" badge) used for app store review and product testing. They are not real people and no personal data of yours is shared with them beyond what any member would see.
  • Amazon Web Services (AWS) — hosting in Canada (Central, ca-central-1) for the website, our self-hosted sign-in service, the PostgreSQL database and private file storage (including photos); transactional email through Amazon Simple Email Service (SES) in Canada (Central); and Amazon Bedrock for photo screening and movie recommendations. Screening requests are processed by Amazon Bedrock in Canada (Central) or the United States (N. Virginia, Ohio or Oregon), depending on the model. For recommendations we send only film titles (with year and type): your favourites, plus titles recently recommended to or dismissed by you so we do not repeat them; or, in a match conversation, both members’ favourites plus titles recently recommended or dismissed in that conversation. Never photos, names or messages. Recommendation requests are processed in the United States. Requests to our website and app pass through Amazon CloudFront and AWS WAF, which protect the service against attacks and abuse. CloudFront receives each request at the AWS network location nearest you, which may be outside Canada, and passes it on encrypted to our servers in Canada. It doesn’t store our pages or your content. AWS WAF checks each request, including your IP address, against known attack patterns and request-rate limits, and may block it. AWS keeps a small sample of checked requests for up to 3 hours so we can see what was blocked; we don’t keep firewall logs.
  • Google LLC & Apple Inc. — optional OAuth sign-in only if you use those buttons
  • Apple Push Notification service (Apple Inc.) and Firebase Cloud Messaging (Google LLC) — deliver phone notifications if you allow them. They receive your phone's notification token and the alert's short text (such as "New message"), never names, message text or photos.
  • ipwho.is, ipapi.co, and geojs.io — IP-to-country lookup for the location check and security emails, tried in turn if one is unavailable (only your IP address is sent)
  • BigDataCloud, OpenStreetMap Nominatim, and geocode.maps.co — reverse geocoding of your device coordinates to country/city, tried in turn if a provider cannot resolve your country. The lookup can receive the latitude and longitude supplied by your device without first rounding them; this is separate from the approximate coordinate stored only when you opt into the 100 km filter.
  • OMDb API — public film metadata (no personal data sent)
  • In-app purchases only: if you subscribe inside our mobile app, your app store processes the payment and shares a subscription status token with us. We do not receive or store your card details. No payments are collected on the website.

A current list of subprocessors is maintained on this page; material changes are announced in-app before they take effect.

Retention and deletion

  • Active account: we keep your data for as long as your account is active.
  • Paused account: your data is preserved but hidden from Discover and Matches until you unpause it.
  • Deletion: you can delete your account from the Your account tab under Account controls at any time. In normal circumstances deletion happens immediately and cannot be undone — there is no grace period and no recovery window. Your login, profile, photos, favourites, likes, matches, messages, verification records, blocks, feedback and device records are removed from our production database when erasure takes effect. If there is an unresolved safety report about your account, the account is closed and hidden from everyone straight away, and the data is erased once moderators finish their review — usually within 7 days. You are told this on screen when it applies. Some records are kept after deletion for the periods described below: the security log, moderation evidence, ban fingerprints, reports about you, legal holds, and backup copies until they expire.
  • Verification anti-abuse data: the submission IP address recorded with a verification request is kept with that request (see below). Rate-limiting counters (which action, when, by which account — no IP) are kept for 7 days, then deleted. Password-reset requests (the email address and time, used only to limit repeat requests) are also kept for 7 days and removed with your account.
  • Messages: messages you exchange with a match are retained for the life of that match. When either party unmatches, the conversation disappears for both members; a private copy is kept for moderators for up to 365 days (longer while a report about it is open) to prevent abuse, then deleted. When either party deletes their account, the conversation is removed as part of that deletion, apart from any moderation copies described here.
  • Verification selfies: your most recently verified selfie is kept while your account exists — it is also your profile photo — so a human moderator can compare it with your next update if a safety concern arises. (Accounts verified before October 2026 may have a second, private pose photo from the earlier two-pose process; it is kept and replaced the same way.) When a newer selfie is verified, the previous one is replaced and deleted shortly after. A capture you start but never submit is deleted after 7 days. A submission still waiting for moderator review expires after 30 days and is then deleted; you can simply verify again. Withdrawn and rejected selfies (and their records) are kept for up to 90 days so we can investigate appeals and repeat abuse, then automatically deleted. Our verification images and records are removed when account erasure takes effect under the deletion rules above. The one exception is a permanently banned account: we keep one-way fingerprints (SHA-256 hashes) of its selfies for 12 months, and of its email address while the ban stands, so the same files or address cannot simply be used again. A fingerprint cannot be turned back into a photo or an email address.
  • Free verification count: we keep a count of the free verification submissions you have used in the current 12-month period (a number and the period start date, no photo), so the free-attempt limit still applies after old selfies are deleted. It is deleted with your account.
  • Moderation evidence: when a message is reported or automatically flagged, or when a conversation ends with an unmatch, a copy of the relevant conversation is archived for moderator review and kept for up to 365 days (longer while a report about it is open), then deleted. This archive is accessible only to moderators and is retained even if the underlying conversation is removed.
  • Reported messages: when you report an individual message, a copy of that message is stored with the report under the same 365-day moderation retention above, so the evidence survives if the sender later deletes or unmatches. The person you reported is not told who reported them.
  • Phone notification tokens: deleted when you sign out on that phone, when Apple or Google tell us the token no longer works (for example after you remove the app), or when you delete your account.
  • Audit and safety logs: the administrative audit log (moderation actions, verification decisions, role grants) is retained for 24 months, then deleted.
  • Security log: the log of account actions (such as blocks, reports, photo removal and account deletion), with the IP address and device/browser type recorded for each (and, for account deletions, a one-way fingerprint of the email address), is kept for 24 months, including after your account is deleted, then deleted.
  • Diagnostics: error traces and technical context attached to bug feedback are kept for 30 days, then deleted; feedback text follows account retention. Grouped issue records are deleted after 30 days without a new report. Network-hash rate-limit counters expire after one hour and are deleted by subsequent reporting or daily cleanup. Diagnostics are used only for troubleshooting.
  • Backups: account erasure removes our live production records and photos. Database backups are kept for a limited period and then deleted automatically, so a deleted account’s data disappears from backups as they expire.
  • AWS exceptional safety copies: the application schedule above covers copies we control. Any image copies held by AWS for exceptional child-safety handling follow the separately disclosed AWS procedures and obligations in “Automated photo screening”.
  • Inactive accounts: if you neither sign in nor open the app for 2 years, we email you about 30 days in advance; if you still don't return, your account and all its data (including verification selfies) are deleted the same way as an in-app deletion. Signing in or opening the app before then cancels it.
  • Legal holds: in the narrow case of an active safety investigation or a legal obligation, we may retain the minimum records necessary to comply, and delete them once the obligation ends.
  • Reports about you: reports filed by other users about your conduct may be retained after your account is deleted for community-safety purposes, without your name or contact details. Closed reports are deleted after 365 days.

An unfinished capture is deleted after 7 days; pending review expires after 30 days; withdrawn and rejected selfies and records are kept up to 90 days. Your latest approved selfie is kept privately while your account exists (even if you remove it from your profile) and replaced shortly after a newer approval. Normal account deletion removes production photos and verification records immediately; an unresolved safety report may delay erasure, usually by no more than 7 days, while the account is hidden. Fingerprints of a permanently banned account’s selfies are kept for 12 months. AWS-held exceptional copies and backup expiry follow the separately disclosed rules.

Your rights

You can access, correct, export, or delete your data from within the app. Open Your account → Controls → Export my data → Prepare download to build a JSON file for data portability. It includes your account email, profile and settings stored in your profile, favourites, likes sent and received, matches, sent and received messages in conversations you can still access, verification records, dated selfie consents, reports you filed, blocks you made, feedback, referrals, and the subscription and extra-verification purchase records stored by us (product, platform where applicable, status, dates, and whether an extra-verification credit has been used). These billing records are not a complete store invoice, charge or renewal history; payment receipts and card details are not included. Photos are referenced by their storage paths, not embedded as image files. This export covers the records listed here, rather than every category of personal information we may hold; contact our Privacy Officer to request other information, including image copies, Our Movie stories, recommendations, device records, conversation-mute settings or separately stored location coordinates. The file is stored in a private storage bucket and handed to you as a single download link that expires about one hour after it is created; the link is only shown to you, and the stored copy is removed by daily cleanup within 24 hours after expiry, or when you request a new export. Because an export can contain private conversation content, we warn you before you download it — keep the file somewhere only you can reach, and delete it when you are done. For the protection of others, the export does not include reports or blocks that other members have made about you, or internal moderation notes and audit-log entries about your account; you can still request access to that information by contacting our Privacy Officer, and we will provide it to the extent it can be disclosed without identifying the reporting member. If you are in Canada, you have the right under PIPEDA (and, in Quebec, under the Act respecting the protection of personal information in the private sector, as amended by Law 25) to access and correct your personal information, to withdraw consent, and to request data portability; you may complain to the Office of the Privacy Commissioner of Canada or, in Quebec, the Commission d'accès à l'information. If you are in EU/UK, you also have the right to restrict processing, object to processing, and lodge a complaint with your national data protection authority. If you are in California, you have the right to know, delete, correct, and opt out of sale/sharing (we do not sell or share personal information for cross-context behavioural advertising). If you are in Illinois or Texas, you have the rights described in BIPA and CUBI in relation to biometric identifiers, including the right to know what we retain and to have it destroyed on request. For any request, contact our Privacy Officer (see Contact below).

EU / UK representative and DSA point of contact

We do not currently maintain a physical establishment in the European Union or United Kingdom. Users in the EU, EEA, or UK may contact us for GDPR/UK GDPR matters, and for matters under the EU Digital Services Act (DSA), at support@bigscreensoulmates.com, which serves as our single point of contact for users and authorities. Where the appointment of an Article 27 GDPR representative or an EU/UK DSA legal representative becomes required, we will appoint one and publish their contact details on this page.

Security

Passwords must be at least 10 characters and include a mix of character types, and new or changed passwords are checked against known breached-password lists (only a partial one-way fingerprint is checked; the password itself isn't sent). We email you when your account is signed in on a new device and when your password is changed. Data in transit uses TLS. Data at rest is stored on encrypted infrastructure with row-level security. Photos are served via short-lived signed URLs. Passwords are hashed with a modern algorithm; we never see them in plaintext. Open Your account → Controls → Sign out of all devices, then confirm with Yes, sign out everywhere. This signs you out everywhere, including here; other devices may take up to a minute, so you can recover access if you signed in on a device you no longer control. We recommend changing your password afterwards. Accounts are limited to three signed-in devices at a time. Keep your sign-in details private and contact us promptly if you believe someone has accessed your account.

Children

The service is 18+ only. We do not knowingly collect data from anyone under 18. In the phone app, we also check the age range Apple or Google provides where the law requires it. Verification photos that appear to show a child are not accepted, and our checks are designed to stop them from becoming a profile photo. If we learn we have data from anyone under 18, we delete it immediately and, where an incident involves child sexual exploitation content, report to NCMEC (US) or the equivalent authority in the applicable jurisdiction as required by law.

International transfers

Big Screen Soulmates is operated from Canada. Our website, sign-in service, database, private file storage and email service are hosted by AWS in Canada (Central). Requests to our website and app first reach Amazon CloudFront at the AWS location nearest you, which may be outside Canada, before continuing to our servers in Canada. Photo-screening requests are processed by Amazon Bedrock in Canada (Central) or the United States, depending on the model. Movie-recommendation requests also go to Amazon Bedrock but are processed in the United States: they carry only film titles with year and type (favourites, plus titles recently recommended or dismissed so we do not repeat them), never photos, names or messages. AWS documents that cross-region traffic stays on its network and is encrypted in transit. Other providers listed above may also process information outside your province or country of residence and may be accessible to law enforcement or national-security authorities under the laws of those jurisdictions. Where transfers occur, we rely on the transfer mechanisms our providers offer (standard contractual clauses, adequacy decisions, or equivalents).

Changes

We may update this policy from time to time. Material changes will be surfaced in the app before they take effect.

Contact

Big Screen Soulmates is operated by Birch Bark Incense Inc., a corporation incorporated under the Canada Business Corporations Act, with its registered office at 67 Third Avenue, Wawa, Ontario, P0S 1K0, Canada.

For privacy questions, or to exercise any of the rights described above, contact our Privacy Officer:

  • Privacy Officer: Brian Harbord
  • Email: support@bigscreensoulmates.com
  • Mail: Birch Bark Incense Inc., Attn: Privacy Officer, PO Box 1321, Wawa, ON P0S 1K0, Canada

You may also submit privacy requests through the Feedback tab inside the app.