Disclaimer
What a scan actually is
A DisclosureProof scan is a deterministic technical check: we load a page in a real browser, look for specific signals (chat-widget scripts, disclosure phrasing, media provenance metadata, AI-content labels), and report what we found. A scan is not a legal opinion, not a certification, and not a substitute for review by a lawyer familiar with your business and the EU AI Act, the NY FAIR News Act, or any other AI-transparency law.
Findings are indicators, not verdicts
Every finding in a report uses one of a small set of neutral states — detected, not detected, or needs manual verification — each carrying a confidence level. We do not use the words "compliant," "violation," or "guarantee" to describe a scanned site, because an automated crawler cannot make that legal determination. Absence of a signal (for example, no machine-readable provenance metadata found on a sampled image) is reported as an absence, never as a violation — only AI-generated content must carry that marking under Article 50(2), and origin cannot always be established automatically.
Automated checks have real limits
Detection is best-effort and can miss things or flag things that need a second look: dynamic content that loads after our capture window, chat widgets with interaction patterns our recipes don't yet cover, disclosure text in a language or phrasing our lexicon doesn't recognize, or a target that blocks automated browsers outright. Where we can't establish a fact with confidence, the report says so explicitly (UNVERIFIED) and offers a short manual check — we do not guess and report a false pass or fail.
No attorney-client relationship
Using DisclosureProof, reading a report, or corresponding with us at hello@disclosureproof.com does not create an attorney-client relationship or any professional-services relationship. We are not a law firm and do not offer legal advice.
The law keeps moving
The EU AI Act's Article 50 guidance, the Commission's Code of Practice, national enforcement structures, and related laws (the NY FAIR News Act, US state AI-transparency statutes, and others) are still being finalized and interpreted as of 2026-07-04. Rule-pack versions are recorded on every scan precisely so a report can be tied to the rules as understood at that time — but "as understood at that time" is not the same as "final legal certainty." Verify current obligations with qualified counsel and official sources before relying on any report.