What is the EU Artificial Intelligence Act? | UpGuard

This document outlines the EU Artificial Intelligence Act, a comprehensive, risk-based regulatory framework for AI across the EU, detailing four risk categories (unacceptable, high, limited, minimal), scope (including non-EU entities impacting EU citizens), enforcement by the European AI Office and national authorities, and significant penalties for non-compliance. It highlights GPAI obligations (documentation, copyright safeguards, security and risk evaluations, and systemic risk considerations), key implementation timelines (initial bans ~Oct 2024, broader requirements within 12–24 months, full enforcement by April 2026), and practical steps for organizational compliance such as risk assessments, transparency, human oversight, governance, monitoring, and staff training.