EU Artificial Intelligence Act Regulations, AI Office Guidelines & GPAI Rules
Track Regulation (EU) 2024/1689 (AI Act) implementation, European AI Office codes of practice for General Purpose AI, high-risk AI standards, and prohibited practice bans.
AI Governance & Emerging Tech Regulatory Landscape & Compliance Overview
Official IntelligencePolicyIndex indexes real-time gazette notifications, policy orders, tariff determinations, and compliance circulars issued across the AI Governance & Emerging Tech sector. Regulatory intelligence is aggregated across central ministries, state regulatory commissions, and statutory authorities.
Statutory Directives
Official Extraordinary Gazettes, S.O. & G.S.R. orders, and executive notifications.
Compliance Mandates
Sector-specific regulatory obligations, licensing norms, and statutory filing guidelines.
Other Verticals in Technology, Digital & Telecommunications
Looking for Weekly Policy & Regulatory Briefings?
Access curated sector roundups, executive summaries, and downloadable PDF intelligence reports.
Official AI Governance & Emerging Tech Gazettes & Notifications
Frequently Asked Questions on AI Governance & Emerging Tech Regulations
How does the EU AI Act classify artificial intelligence systems?
Regulation (EU) 2024/1689 establishes a risk-tiered framework: Unacceptable Risk (banned practices such as cognitive behavioural manipulation and social scoring), High Risk (regulated critical infrastructure, employment, biometric systems subject to strict CE-marking and risk management), Specific Transparency Risk (chatbots, deepfakes), and Minimal Risk.
What requirements apply to General Purpose AI (GPAI) model providers?
GPAI providers must maintain technical documentation, comply with EU copyright law, publish training content summaries, and, for models posing systemic risks (>10^25 FLOPs), perform model evaluations, adversarial testing, and report serious incidents to the European AI Office.
What are AI regulatory sandboxes under the AI Act?
Article 57 of the AI Act requires each Member State to establish at least one AI regulatory sandbox by August 2026, providing a controlled environment for developing, testing, and validating innovative AI systems under regulatory supervision. Sandboxes must offer priority access for SMEs and start-ups, and national authorities must provide guidance on compliance expectations and inform the European AI Office of outcomes.
What AI practices are prohibited under the EU AI Act?
Article 5 prohibits AI systems that deploy subliminal or manipulative techniques causing harm, exploit vulnerabilities of specific groups (age, disability), perform social scoring by public authorities, conduct untargeted facial recognition scraping, engage in emotion recognition in workplaces and education (with limited exceptions), and carry out biometric categorisation based on sensitive attributes such as race, religion, or sexual orientation.