AI Transparency Policy

AI may read product-label images, suggest product fields and categories, detect possible duplicates, normalise brand information, identify spam or policy risks, group recurring themes, and generate summaries of reviews and defect reports.

Automated tools support moderation and organisation; they do not transform FYDERIO into a product tester, certifier, manufacturer, seller, or professional adviser.

AI can misunderstand images, context, language, sarcasm, product variants, uncommon failures, and conflicting reports. Outputs can contain omissions, outdated information, unsupported inferences, or errors even when presented fluently.

The quality of an insight depends on the quantity, diversity, accuracy, and recency of the underlying information. A small or unrepresentative sample can create a misleading impression.

AI summaries and classifications are informational aids, not verified facts. Read the underlying community contributions, product label, manufacturer instructions, and official safety information before relying on them.

FYDERIO may update or withdraw an insight as new information becomes available or quality controls identify a problem.

A risk score is a platform indicator derived from available community reports, including the reported frequency and severity of defects. It is relative to the data available to FYDERIO and is not a statistical prediction of failure for every unit.

A low score does not prove safety or reliability. A high score does not establish that a product is legally defective, unsafe, or subject to recall. Scores may change as reports are added, corrected, moderated, or reweighted.

You remain responsible for purchase, use, repair, disposal, and safety decisions. Check model and batch details, follow official instructions, monitor recall notices, and stop using a product when you reasonably suspect danger.

FYDERIO’s AI output is not medical, legal, engineering, financial, laboratory, regulatory, or other professional advice. For urgent safety concerns, contact the manufacturer, retailer, competent authority, emergency service, or an appropriately qualified professional.

Users may report an AI insight that appears inaccurate, unsafe, biased, or misleading. FYDERIO may review the source material, revise the output, adjust systems, or add context. Moderation appeals involving automation may receive human review where appropriate.