<ul data-eligibleForWebStory="false"><li>The European Union is introducing new AI rules that focus on copyright protections, transparency, and public safety, which will initially be voluntary for major AI companies before becoming mandatory in August 2026.</li><li>Companies like Google, Meta, and OpenAI will be required to promise not to pirate materials for AI training and share detailed information about their training data, including disclosing data sources and energy consumption.</li><li>The rules also require AI companies to respect paywalls and robots.txt instructions and provide guidance on safety measures, including monitoring for potential harms and reporting serious incidents within specific timelines.</li><li>The new rules aim to address concerns about AI innovation and ensure transparency and compliance, with potential penalties for non-compliance including fines of up to 7% of a company’s annual sales.</li></ul>

Everything tech giants will hate about the EU’s new AI rules

Discover more