Regulation
27 analyses
Trump confirmed on June 5 negotiations for the federal government to acquire shares in OpenAI through a sovereign fund. Senator Bernie Sanders responds with a competing proposal demanding a mandatory 50% stake in OpenAI, Anthropic, and xAI.
Text signed on June 2 establishes an AI Cybersecurity Clearinghouse, grants 60 days to agencies to publish the security framework, and assigns NIST the task of classifying covered frontier models.
OpenAI has published a formal framework aligned with the EU AI Act and California's TFAIA, defining systemic risk as any model that causes more than 50 deaths or $1 billion in damages from a single incident.
The House approved the measure 110 to 0. Pritzker is expected to sign it this week. Illinois will require an annual independent audit of safety protocols in frontier models, ahead of the European federal benchmark.
The US President cancelled on Thursday the signing ceremony of an executive order that would create a federal safety review mechanism for AI models before their launch, citing risks of slowing down America's advantage over China.
President Donald Trump is expected to sign an executive order this week prompting AI labs to grant government access to models 90 days prior to their release. The order also revises federal sharing of cyber threat information.
Guidelines published on May 19 aim to unlock companies stalled by classification uncertainty. The market is estimated to reach up to €38 billion in compliance by 2030, with initial costs of up to US$15 million for large corporations.
Political agreement between the Council and Parliament postpones the August 2026 deadline and opens a 19-month window for compliance, but fines and the extraterritoriality clause remain intact.
In August 2026, the EU AI Act imposes its first binding obligations for high-risk systems. Meanwhile, four US states have AI laws in effect that clash with the federal strategy of the Trump administration. For companies, compliance now means managing conflicting jurisdictions.
The Illinois Biometric Information Privacy Act (BIPA) has generated over $2.5 billion in settlements since 2008 and continues to produce record litigation. In 2025, Texas and Washington expanded similar legislation. In Europe, the AI Act classifies remote identification biometric systems as high risk. The proliferation of facial recognition cameras and voice authentication in the corporate environment exposes companies to increasing legal risks.
Since 2018, GDPR has accumulated €7.1 billion in fines, with 60% of the total issued after January 2023 alone. In August 2026, the EU AI Act adds a second layer: penalties of up to €35 million or 7% of global revenue. Organisations processing data with AI face dual and simultaneous regulatory exposure.
China has built the world's most detailed and layered AI regulatory framework. With three main regulations already in effect and at least two more in preparation, the country requires registration of foundational models, pre-launch security assessments, and control of AI-generated content. Companies with operations or users in China need to map compliance across multiple fronts simultaneously.