AI Analysis Policy

1. How the analysis works

Sentava analyses a listing using several independent signals:

• listing content — consistency of the title and description and typical risk signals,
• seller — publicly available information about the account, where available,
• photos — automatic analysis of the photos attached to the listing,
• transaction terms — payment, delivery and other conditions stated in the listing,
• price — comparison of the asking price with verified information about the exact product, including an authentic reference price when it can be reliably established.

Sentava does not assess product authenticity using prices of other, similar marketplace listings.

2. What the result is NOT

• it is not a guarantee that a transaction will be safe,
• it is not a certificate of product authenticity,
• it is not an accusation or confirmation that a seller is a fraudster,
• it is not legal, financial or expert appraisal advice.

The result is guidance meant to support — not replace — your own decision.

3. Limitations and possible errors

Automated analysis can be wrong. It may flag elevated risk where there is none, or miss a genuine threat. No automated system offers complete certainty.

4. Dependence on available data

The accuracy of a result depends on the data we can reach at the moment of the Check. If an authentic reference price cannot be reliably established, we say so clearly instead of showing an estimate. Missing data is never treated as evidence of fraud.

5. Photo analysis

Photo analysis detects visual signals that may suggest elevated risk, such as photos that do not match the description or traits typical of counterfeits. It is not a certification of authenticity.

6. External AI provider

To run an analysis, the listing content and its public photos are sent to an external AI model provider (Lovable AI Gateway, using Google Gemini models). Details about data processing are described in our Privacy Policy.