1. Who we are
Nibavo is operated by Vladyslav Kravchenko, an individual developer based in Ukraine (“Nibavo”, “we”, “us”). We are responsible for the processing described here.
Privacy and support email: support@rialza.app. Please include “Nibavo” in the subject. This shared mailbox also supports RIALZA; identify which application your request concerns.
This policy covers the Nibavo mobile application and our Nibavo website. It explains information processed on your device as well as information processed by our service providers.
2. What Nibavo does
Nibavo helps you photograph food or packaging, understand available label information and compare it with preferences you choose. Cloud AI analysis is optional and requires your permission. You can keep photographs locally without sending them for AI analysis.
We use an automatically generated technical account for cloud requests. It is associated with a persistent identifier, not necessarily with your name or email. This identifier is pseudonymous, not a promise that your activity is unidentifiable.
3. Information we process
| Category | Examples and processing |
|---|---|
| Your optional food profile | Allergies, ingredients you avoid, eating style, dietary practices, fasting choices, nutrition goals and text you enter. The present app stores the profile on your device and evaluates profile matches locally. The profile fields are not included in the implemented AI-analysis request. |
| Photos and label information | Photos you take or select, product/brand text, ingredients, nutrition, packaging claims and other visible content. Selected photos are processed remotely when you permit AI analysis; the whole selected image can be visible to the provider, not just words on the label. |
| Local library and preferences | Saved photos/results, earlier image versions retained as evidence, history, favorites, reactions, corrections, unfinished work, language/settings and a local adult-eligibility declaration. Our persistent user-facing library is local; processing an individual scan still creates temporary cloud content and technical records. |
| Cloud and security records | Technical account ID, app/attestation tokens, scan and image identifiers, dimensions, file hashes, timestamps, permission versions, request status, usage limits, operational errors and deletion records. Google and platform services may also process IP addresses, user-agent/device information and security metadata. |
| Support correspondence | Your email address, message and information you choose to attach. Please avoid sending medical records, passwords or identity documents unless a specific, proportionate verification need has been explained. |
| Website requests | Technical information required to deliver and protect the website, potentially including IP address and request/device metadata at the hosting or network provider. The current website has no sign-in, newsletter or payment form. |
Allergies may reveal health information. Religious dietary choices may reveal religious beliefs. We treat these optional profile details as sensitive even when processing happens on your device. A product photograph is not necessarily sensitive, but people, addresses, prescriptions or other details in the frame can make it personal or sensitive. Check the frame before uploading.
4. Why we use information
We process information to provide requested local features and AI analysis, display and save results, apply your chosen profile, maintain continuity of pending work, secure the service, enforce fair-use limits, handle support/privacy requests and meet applicable legal obligations.
Where the EEA or UK data-protection framework applies, our bases are: consent for optional cloud-photo processing; explicit consent for processing optional profile information revealing health or religious beliefs and, where this is the applicable special-category condition, sensitive information in product photos you choose to send; performance of the requested service for necessary service operations; legitimate interests in proportionate security, abuse prevention and support administration; and legal obligations where applicable. These bases do not make every optional use necessary. You may decline optional processing and withdraw permission for future processing.
We request product/packaging photos, not medical documents, prescriptions, another person’s profile or personal records. Please exclude those materials and unnecessary personal details. We do not request clinical records through ordinary support email; if resolving an inquiry genuinely requires sensitive information, we first explain the limited information needed and establish an appropriate basis and suitable channel. Unrequested sensitive material is minimized and handled only to the extent lawfully necessary, not reused for personalization or AI analysis by default.
We do not sell or rent your personal information or consumer health data, disclose it for cross-context targeted advertising, or use sensitive profile details for advertising. We do not use your photos or profile to train our own AI models. These are purpose limits, not a claim that no providers receive technical information needed to run the service.
5. Cloud AI and service providers
Google/Firebase and Google Cloud provide the technical account, app-integrity verification, temporary image storage, job/result handling, server execution and website hosting. Apple and Google platform services can participate in device/app attestation and operating-system permissions. OpenAI receives the selected prepared images and associated image context to interpret the product or label. We do not intentionally include your food-profile fields, full local library or personal reactions in that request.
Before transfer, upload images are re-encoded without EXIF metadata. That does not remove information visibly photographed, and locally retained originals may still contain their original metadata.
We request OpenAI processing with response storage disabled (store:false). The store:false setting does not establish Zero Data Retention or prevent all provider retention. OpenAI states that API inputs and outputs are excluded from model training unless a customer opts into data sharing. Under standard API controls, abuse-monitoring content may be retained for up to 30 days, with documented safety/legal exceptions and account-specific controls. See OpenAI’s API data controls. Our own deletion process cannot instantly recall a request already transmitted to a provider.
We also use the owner-confirmed shared RIALZA support mailbox to receive Nibavo inquiries. Our email service provider processes those messages for delivery, storage and security. Information may be disclosed where legally required, to protect people and the service, or in a lawful business transition with appropriate notice and safeguards. We do not publish submitted photos to a community or public product database as part of the present service.
6. Retention
Different copies have different lifetimes. Expiry of access and physical deletion are not the same event.
| Information | Retention and limitations |
|---|---|
| Local profile, library and results | Kept for your use without a general automatic age-based expiry. Removing or replacing a photo in a product’s current photo set can retain an earlier version as evidence for a previous result. Delete the entire saved product/photo session to request removal of its local current and archived copies; interrupted file cleanup may complete later. The profile and analysis records use private app storage excluded from normal app backup; photo-library files have different OS backup behavior. Your own gallery copies and device backups are separate. |
| Temporary cloud image copies | Removal is attempted after processing. Unfinished/orphan uploads are scheduled for cleanup under a 24-hour application-expiry policy and an object-lifecycle rule. Processing failures, cleanup backlogs and infrastructure timing can delay removal; this is not an absolute 24-hour physical-erasure guarantee. |
| Undelivered results and ordinary job metadata | Temporary delivery uses a seven-day application-expiry window; acknowledgement can remove a result earlier. Scheduled cleanup handles remaining records asynchronously. This is not a maximum for all security/provider records. |
| Security, consent, deletion and anti-replay records | Minimal owner-bound permission, deletion and anti-replay records are kept while the technical account remains active and while a deletion request is unresolved. After account deletion is confirmed, records protecting against stale work become eligible for removal after a 24-hour protection period and a fresh check that the account no longer exists. Cleanup is asynchronous and remains pending if confirmation or removal fails. A completed deletion-status receipt contains no account identifier, profile, photos or results; it expires 30 days after completion and is removed asynchronously. An unresolved request is not silently marked complete or discarded just because time has passed. These limited records serve permission, fraud prevention and prevention of deleted work being re-created; they do not justify retaining all label content indefinitely. |
| Infrastructure logs and recovery copies | Current project logging buckets have 30-day ordinary-log and 400-day required-audit-log retention. Whether a particular event is recorded depends on its category. The present image bucket has object versioning and soft delete disabled; Firestore retains recent versions for a one-hour window even without point-in-time recovery enabled. These settings do not eliminate provider security records or every independently created backup/export. |
| Authentication/provider records | Follow the relevant deletion process and provider’s retention rules. Firebase describes residual authentication information as removed from live/backup systems within up to 180 days after account deletion is initiated. OpenAI handling is described above. |
| Support correspondence | Kept while resolving your inquiry and for proportionate follow-up, security or legal needs; request deletion or ask us about retained correspondence. Any longer legal hold must have a specific justification. |
There is no general cross-device recovery promise. Uninstalling the app does not by itself submit a cloud-deletion request. Restoring a device backup may restore copies outside a deletion completed in the live app.
7. Your controls and requests
You can choose whether to add a profile or send photos to AI. You can edit/delete your profile, delete an entire saved product/photo session, and stop future photo uploads from the product’s menu. Removing or replacing just one photo from the current set may archive its earlier version for past-result evidence; that action is not erasure of every copy. Withdrawal does not automatically erase earlier results or all prior technical records. When offline, local withdrawal takes effect locally, while server requests must wait for a connection; processing already transmitted cannot be undone.
For the scope of local and account deletion and the available request routes, see Delete your Nibavo data.
For access, correction, deletion, a copy of information, consent withdrawal or another privacy request, email support@rialza.app with subject “Nibavo privacy request”. Explain the requested action and whether you still have the app/device. We may need proportionate ownership verification for a pseudonymous account. Do not send account passwords, tokens, medical records or other unnecessary documents. If we cannot reliably identify an old anonymous installation, we will explain the limits and reasonable available alternatives rather than give another person’s data to you.
Depending on your location and applicable law, you may also have rights to object, restrict processing, receive a portable copy, withdraw consent, appeal a refusal or complain to a data-protection authority. Withdrawal does not invalidate earlier lawful processing. We respond within applicable legal periods, explain any lawful extension or refusal, and do not unlawfully penalize you for exercising your rights. Details for consumer health data are in our Consumer Health Data Privacy Policy.
8. International processing
The operator is based in Ukraine. The currently configured application backend, temporary image bucket and Firestore database are in Google’s us-central1 region in the United States. Authentication, attestation, hosting, support and AI providers may process information in other countries. A selected server region is not a promise of wholly local or EEA-only processing.
Our initial app launch is intended for users in the United States. Processing may take place in other countries as described above. Before offering the app in additional markets, we review the notices, rights and international-transfer arrangements required for those markets. Contact us for information relevant to your request. A privacy-policy acceptance is not a substitute for a required transfer mechanism.
9. Security
We use measures appropriate to the service, including private app storage, authenticated/owner-bound service operations, app-integrity checks, access rules and transport security. Nibavo is not an end-to-end encrypted AI service: selected content must be available to the processing providers. No storage or transfer method can be guaranteed perfectly secure. Keep your device and its account secure.
10. Children
Nibavo is intended for adults aged 18 or older and is not directed to children. Do not enter a child’s medical profile or upload a child’s personal information for analysis. If you believe a child has provided personal information, contact us so that we can investigate and take appropriate action. An age statement does not replace our obligations where we become aware of children’s data.
11. Website cookies and tracking
The present website does not implement optional analytics, advertising tags, tracking pixels, third-party embeds or authored browser-storage/cookie functionality. Hosting, network and security providers may still process technical requests. If optional tracking is introduced, we will update this notice and provide any consent/refusal controls required before it is activated.
12. Changes
We display the effective date of this policy and make material changes clear. When a change requires a new choice or consent, we will obtain it before using information for that new purpose. Merely changing this page will not silently enable additional optional processing.