From 7ff3da2c0669e2e28f4a3eb555aa1d91fa091a7b Mon Sep 17 00:00:00 2001 From: "Ioannis L." <44038245+blitzcrieg1@users.noreply.github.com> Date: Fri, 6 Nov 2026 12:52:22 +0200 Subject: [PATCH] docs(compliance): transparency is Art. 50 of the AI Act, not Art. 52 The deployer checklist headed its transparency section "Art. 52". In the final text Art. 52 is the procedure for classifying general-purpose models with systemic risk, and transparency is Art. 50. A buyer's counsel checking the citation would have found the wrong article, in a document whose job is to be checked. The steps under the heading are also not what Art. 50 requires: it binds providers, and deployers only in the cases it lists. A note now says so, and points to Art. 26(7) for informing workers about a high-risk system. Co-Authored-By: Claude Opus 5.5 --- docs/compliance/ai-act-deployer-checklist.md | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/docs/compliance/ai-act-deployer-checklist.md b/docs/compliance/ai-act-deployer-checklist.md index 3ed55f3f..efa34622 100644 --- a/docs/compliance/ai-act-deployer-checklist.md +++ b/docs/compliance/ai-act-deployer-checklist.md @@ -40,7 +40,13 @@ that nothing happened outside the monitored boundary.* --- -## Art. 52 — Transparency (user-facing) +## Art. 50: Transparency (user-facing) + +Art. 50 binds providers of systems that interact with people or generate content, +and deployers only in the cases it lists. The steps below are good practice for +telling developers about monitoring, not a list of what Art. 50 requires. Where a +high-risk system is used at work, Art. 26(7) separately requires employers to +inform workers' representatives and affected workers first. | Step | Action | Agentmetry hook | |------|--------|-----------------|