Privacy Policy

Revision dated 20 October 2026.

Controller: Dmitri Lõzlov, an individual. Address: Estonia, Tallinn, Mustakivi tee 25, 13912. Contact: dmitrilyzlov@icloud.com.

Manage consent · Delete account

Reports, blocks and recipe review

The administrator reviews new and edited community recipes, including photos, before publication. Reports store the publication and reporter identifiers, reason, explanation and time. Only the administrator receives reports; the author does not receive the reporter identity. Open reports are kept until review; resolved reports are deleted within 30 days of the decision. Blocks are kept until removed or a related account is deleted. Account deletion removes related reports and blocks; backups expire within the stated 30 days. The legal basis is the legitimate interest in service safety (GDPR Article 6(1)(f)). Do not include health information or other people's personal data in an explanation.

Scope and controller

This policy describes how personal data are processed in Honest Plate at honestplate.eu and in the app that uses this service. The controller is the owner identified above, an individual in Estonia. This policy takes effect on 20 October 2026. For questions about your data or to exercise your rights, contact the email address above.

What data we receive and where they come from

You provide your profile name, food, weight, goal and exercise entries, personal foods, recipes, photos, dietary preferences and restrictions, and language and time settings. The sign-in service provides an identifier, verified email address and available name. Connecting a coach creates an account relationship; publishing a recipe stores the chosen author name and reactions. Servers also receive technical connection information. Some weight, food and dietary restriction entries may reveal health information.

Purposes and legal bases

Registration, sign-in, the features you request and synchronization of ordinary data rely on Article 6(1)(b) GDPR. Diary data that may reveal health information are processed with separate explicit consent under Articles 6(1)(a) and 9(2)(a). AI transfers, coach access and publication are separate choices. Security, abuse prevention, limited logs and preventing deleted accounts from being restored rely on legitimate interests under Article 6(1)(f), using only the minimum necessary data. Mandatory responses to requests and compliance with applicable legal requirements rely on Article 6(1)(c).

Your choices and withdrawing consent

Before you use the diary, the app asks you to separately confirm the processing of your data, including health information. The account identifier, date and consent revision are recorded; viewing this policy does not itself constitute consent. Without confirmation, the server does not accept new diary data or personal photos. Settings are available through the “Manage consent” link on this page. Each photo transfer or recipe request to OpenAI requires a separate confirmation; permission does not carry over to the next request. Before sending, you can uncheck the consent box or close the form. To stop future AI transfers, simply do not confirm new requests; the diary and manual food entry remain available. This does not undo previous transfers, which remain subject to the OpenAI retention terms below. You can withdraw diary consent through “Delete account” or by contacting the owner: diary processing stops and entries are deleted, subject to the backup terms below. You can export your data before deletion. Withdrawal does not affect the lawfulness of processing that took place before it.

External services and recipients

Sites/OpenAI provides website hosting and operation; Cloudflare provides server code execution, the D1 database, R2 file storage and network security. Supabase Auth handles registration, sign-in, sessions and account deletion; Supabase also runs maintenance jobs. The Supabase project region is eu-central-1 (Frankfurt), but support and certain service operations may take place in other countries. These providers and their authorized subprocessors receive data as needed for their tasks. The owner has administrative access for support, troubleshooting, backups and handling requests.

Sign-in and emails

The current configuration supports email and Google sign-in; existing users may also use ChatGPT sign-in. Apple sign-in is currently disabled. Social sign-in does not pass your Google or ChatGPT password to Honest Plate. Email sign-in passwords are handled by Supabase Auth. Confirmation and recovery emails are sent by the authentication service’s email infrastructure. Google and OpenAI independently process data under their own policies when you use their accounts.

Sending photos to OpenAI

Only after your separate confirmation and after you start analysis does the app send the selected food photo, entered description, total dish weight and response language to the OpenAI API. The code for this request does not add your name, email address or diary, but the photo and description themselves may contain personal information: avoid including other people, documents or unnecessary data. A photo used for recognition is not stored as a permanent recipe photo; you can save the result you approve to your diary.

AI recipe suggestions

This optional feature works only when the service is configured. After separate consent, the app sends OpenAI the names, ingredients and steps of up to 20 of your own recipes, up to 100 ingredient options, selected allergen exclusions, preferences and response language. This request does not include photos, your profile, diary entries, body weight or goals. Saving a suggested recipe is a separate action. The app sets store:false for both types of AI request.

Retention by the OpenAI API

Under OpenAI’s published terms, API data are not used for model training by default unless you separately opt in to data sharing. store:false disables storing the response as API history, but does not mean zero retention: abuse monitoring logs may be retained for up to 30 days by default, or longer where the provider’s terms allow this for legal or security needs. Separately, prompt caching may retain an encrypted internal representation of data for up to 24 hours; store:false does not exclude this cache. Images are subject to exceptions for prohibited-content checks. Honest Plate does not claim Zero Data Retention or Modified Abuse Monitoring without confirmation from the provider. Deleting your Honest Plate account does not mean that data already transferred are immediately deleted from OpenAI’s internal service systems.

Coaches, publication and the shared catalog

A coach can access the food entries, weight and goals provided by the app only after a confirmed connection. You can remove the connection in “Mentorship”. Publishing a recipe requires a separate confirmation: the recipe, photo and chosen author name become available to other users. You can unpublish it, but copies saved by others are not automatically withdrawn. Eligible user-created foods are sent to the administrator for review for the shared catalog; the entry form explains this. Do not include personal information in a food name.

International transfers

Processing is not limited to Estonia or the EEA. The Cloudflare network, Sites/OpenAI, the OpenAI API, and Supabase’s international operations and subprocessors may process data in countries including the United States; Supabase Pte. Ltd. is registered in Singapore. The European region of the primary Supabase database does not guarantee that all service data remain in Europe. Supabase’s DPA has been incorporated into its Terms of Service since 1 August 2026, as confirmed by support. Processing of personal data of visitors to the published website is covered by section 3.2 of the ChatGPT Sites terms and the DPA incorporated into them; for a controller in Estonia, the Sites DPA contracting party is OpenAI Ireland Ltd. These agreements provide international transfer safeguards, including the European Commission’s Standard Contractual Clauses (SCCs) or an applicable adequacy decision. The OpenAI API, Sites and Supabase terms are linked below; the hosting DPA does not cover the content of conversations with ChatGPT used to create the site. A DPA does not mean zero retention or processing only within the EEA. You can ask the owner for information about safeguards. AI consent does not replace safeguards for international transfers.

Device storage and cookies

For offline use, the diary and pending changes are stored in IndexedDB; settings and consent confirmation may be stored in the browser. Cookies are used for authentication and sign-in security. Normal sign-out keeps the offline copy. When you delete your account, the app clears that copy on the device used for deletion; other devices must connect to the internet to receive the access revocation and cleanup. Copies on offline devices and in previously exported files must be deleted separately. The app’s code does not include advertising trackers; necessary storage technologies support the features you choose.

Automated calculations and AI

Calories, macronutrients and suggestions are calculated automatically, but you review the result and decide whether to save it. The app does not make solely automated decisions that have legal or similarly significant effects. Photo and recipe estimates may be wrong and do not confirm the absence of allergens.

Your rights and how to make a request

You may request access to and a copy of your data, rectification, erasure, restriction of processing, portability where applicable, and withdrawal of consent. You may object to processing based on legitimate interests. Export and deletion are available in the app; send other requests to dmitrilyzlov@icloud.com. Where needed, we will ask for proportionate identity verification. We normally respond within one month; if a lawful extension is needed, we will explain the reason and the new deadline within that month. You may complain to Andmekaitse Inspektsioon (aki.ee), or the supervisory authority in the country of your habitual residence, work or the alleged infringement.

Requests, limitations and changes

Correspondence about data is retained for as long as necessary to fulfill the request and demonstrate its completion; in a dispute, it is retained to resolve the dispute and defend rights for the applicable periods. It is separate from API logs. Do not send unnecessary medical documents. The service is intended for independent diary use; if you have not reached the age at which you can independently consent under applicable law, contact the owner before entering personal data. If the purposes of processing change materially, we will update this page and request new consent where necessary.

Primary data and account deletion

Profile and diary data are retained while the account is in use and until you delete the entries or account; inactive accounts are not currently deleted automatically. “Account → Delete account” requires you to verify your identity again. Once a request is accepted, access is blocked and working data, photos, publications, relationships and the Supabase Auth account are removed. If cleanup temporarily fails, it is retried every five minutes; the result shows when it is complete. The result receipt is retained for up to 30 days. This does not delete your Google or ChatGPT accounts.

Backups — 30 days

A private server backup of user data and associated photos is created daily. It is retained for no more than 30 days from the backup date; after that, download and normal restoration are blocked. Physical cleanup runs every five minutes; infrastructure failures may delay it until a successful retry. Backups created before account deletion may contain earlier data until that period expires. During restoration, the current deletion ledger excludes those accounts and photos before access is reopened. The owner’s service archives in the managed folder are also removed when they expire; the local job requires the computer to be running. This rule does not manage users’ personal exports.

Application logs — 7 and 30 days

API errors are retained for up to 7 days; access denials and rate limits for up to 30 days. The recorded fields are a random event identifier, time, category, fixed operation group, HTTP method and result code. Request text, diary entries, photos, names, email addresses, IP addresses, passwords and tokens are not written to this table. Expired entries are removed by scheduled maintenance; cleanup is retried after failures. This rule does not describe providers’ separate infrastructure logs.

Access revocation and consent records

Minimal opaque identifiers of deleted accounts and hashes of former identities are stored separately for as long as needed to block old sessions and prevent deleted data from being restored. They are not anonymous and contain no diary entries, photos, names or email addresses. They are not covered by the 7/30-day log retention periods. Consent confirmation is kept until account deletion; fresh consent is requested after restoring a backup.

Separate infrastructure retention periods

Sites/Cloudflare may keep network logs containing IP addresses, URLs, times, headers and request results. Cloudflare’s documentation gives standard Workers Logs retention as 3 or 7 days, and D1 Time Travel restoration history as 7 or 30 days, depending on the plan. These are general product rules, not confirmation of the settings, exports or internal copies for our website in Sites. As of 20 October 2026, the Supabase project is confirmed to use the Free plan and eu-central-1 (Frankfurt) region. Published retention for API and Database logs on Free is one day; this does not describe all of the provider’s internal logs. The dashboard does not provide scheduled backups or point-in-time recovery on this plan. This does not prove the absence of internal copies: separate Supabase guidance describes creating up to seven daily backups of Free projects. Support has been asked to clarify the exact retention of those copies, Auth logs and residual data after deletion. The working Supabase Auth account is deleted when the account is deleted in the app. Sites’ internal retention periods also require provider confirmation. Service data may be retained for security, legal compliance and operations under providers’ terms; a single deletion deadline for all such copies has not been confirmed.