n8n AI Act Labeling Workflow: Slack Approval + Postgres

Turn EU AI Act disclosures into an approval-ready pipeline—automatically

This n8n AI Act labeling workflow collects AI-generated assets, classifies them for EU AI Act Article 50 disclosure, routes each item through a human approval gate (with optional Slack notification), and records decisions plus a registry snapshot in Postgres—with optional label burning into media when applicable.

What this workflow does

• Receives assets via an n8n Form trigger or by calling another workflow using Execute Workflow.
• Classifies for EU AI Act Article 50 by hashing the prompt, applying Article 50 rules, and generating a language-specific disclosure sentence and manifest metadata.
• Applies disclosure based on asset type:
  • Text: appends the disclosure sentence when disclosure is required.
  • Media (images/videos/audio): saves the incoming file and attempts to burn in the disclosure label using n8n’s image editing or an HTTP media-labeling service; if labeling fails, it falls back to non–burn-in handling.
• Stores traceability in Postgres, including asset manifests and an approval link, then optionally posts a review request to a Slack channel.
• Waits for human review: the reviewer can approve, apply an editorial exception (text only), or return the asset.
• Creates audit integrity by recording decisions in a hash-chained Postgres ledger and snapshots an AI-systems registry by scanning workflows via the n8n API.

Use cases

• SaaS teams needing human-in-the-loop EU AI Act disclosure workflows for generated content.
• Automation engineers building compliant review pipelines for AI marketing assets, copy, or media.
• Operators who must maintain regulatory traceability with Postgres-backed decision logs and registry snapshots.

Technical details

• Core nodes/logic: if, set, code, wait, slack (as indicated by the workflow stack).
• Integrations: Slack for review requests, Postgres for asset storage, decisions ledger, and registry snapshots, plus n8n API for workflow scanning.