Mirror Reader AI Disclaimer

Effective Date: July 16, 2026

Mirror Reader uses AI-assisted or automated systems for catalog translations, generated covers, preparation and translation of user-provided texts, text alignment, and optional English pronunciation feedback. This notice supplements the Terms of Use and Privacy Policy.

Generated and translated text

AI output can contain mistranslations, omissions, invented details, alignment errors, inconsistent style, outdated wording, or culturally inaccurate language. Output is provided for personal reading and language learning. It is not an official, certified, academic, publisher-approved, legal, medical, or professional translation.

Do not rely on Mirror Reader output as the sole basis for a high-stakes decision. Compare important passages with the source and obtain qualified human review where appropriate.

User-provided texts

Creator users can import books, documents, articles, notes, study materials, and other supported texts. When a user requests cloud preparation or translation, needed fragments and instructions can be sent through an AI routing intermediary to a selected or automatically routed model provider.

Different providers have different policies for retention, training, evaluation, improvement, safety review, and deletion. Mirror Reader does not promise that every request is Zero Data Retention. Do not submit sensitive, confidential, protected, or third-party content unless you have the rights and understand the provider flow.

Pronunciation feedback

Eligible users learning English can optionally record a short word attempt. The recording leaves the device and is processed in parallel: a third-party cloud speech-analysis provider estimates pronunciation accuracy, while a separate Mirror Reader-operated phonetic-analysis service estimates the sounds that were actually spoken. These systems can disagree or be wrong because of accent, microphone quality, noise, speech variation, model limitations, or network conditions.

Pronunciation results are learning aids, not medical, clinical, speech-therapy, accessibility, employment, academic, or certification assessments. They must not be used to diagnose a condition or make a high-stakes decision.

Audio is processed transiently as described in the Privacy Policy. If the Mirror Reader-operated service is unavailable, Mirror Reader can return cloud-provider feedback without the extra detected-sound analysis.

Covers and catalog material

Generated covers are artistic interpretations, not historical evidence or official publisher artwork. Public-domain status can differ by country, edition, translation, illustration, and source. An underlying public-domain text does not make every modern translation, introduction, annotation, recording, or cover public domain.

Reporting a problem

Report a translation, alignment, pronunciation, rights, or content problem to [email protected]. Include the title or text, language pair, relevant excerpt or screenshot, and a description of the issue. Do not include credentials or unnecessary sensitive information.

Mirror Reader may update models, providers, prompts, generated output, and this notice as the Service changes.