Particle.news
Download on the App Store
10 ARTICLES
·
2mo ago

EU Issues Final AI Code of Practice Ahead of August Rule Rollout

Developed by 13 independent experts after broad consultation, the voluntary framework outlines transparency, copyright and safety measures for general-purpose models set to face binding rules on August 2.

European Union flags fly outside the European Commission in Brussels, Belgium November 8, 2023. REUTERS/Yves Herman/File Photo
FILE - The European Union flag stands inside the atrium at the European Council building in Brussels, June 17, 2024. (AP Photo/Omar Havana, file)
European Union flags wave outside the European Commission building in Brussels, where officials unveiled a new code of practice on general-purpose AI. The voluntary guidelines aim to enforce transparency, copyright protections, and safety standards for advanced AI systems like ChatGPT.
The EU has come under fierce pressure to delay enforcing its landmark AI law

Overview

  • The European Commission published the final voluntary Code of Practice on July 10 to help AI companies align with the EU AI Act’s general-purpose model obligations.
  • The code was drafted by 13 independent experts following input from over 1,000 stakeholders and takes effect alongside the AI Act’s compliance deadline on August 2.
  • It requires providers to detail training data sources, adhere to EU copyright law and report serious incidents to the EU AI Office within five to ten days.
  • Companies that sign the code will benefit from reduced administrative burdens and increased legal certainty compared to those proving compliance by other means.
  • French startup Mistral has already committed to the framework, while OpenAI, Google and Meta continue reviewing the guidance ahead of enforcement beginning in August 2026 for new models and August 2027 for existing ones.