AI Regulation: EU AI Act, US States, UK and Korea — 2026-10-29
State legislatures in the US have enacted a record 115 AI laws in 2026, with Illinois introducing the first mandatory third-party audits for frontier model developers. In Europe, the first harmonized standard for high-risk AI systems (EN 18286:2026) has been published, while in Korea, the Federation of Korean Industries formally requested 19 regulatory clarifications to the AI Basic Act to resolve ambiguities regarding internal corporate AI use.
AI Regulation: EU AI Act, US States, UK and Korea — 2026-10-29
Top developments
Record 115 State AI Laws Enacted in US as Illinois Mandates Audits
State legislatures have passed a record 115 AI-related laws across 30 states in 2026, representing a 58% increase from the previous year. Illinois became the first US state to mandate annual independent third-party audits for the largest AI developers, aiming to fill the governance vacuum left by stalled federal legislation. These laws include new chatbot safety requirements and definitions of personhood, signaling a shift toward aggressive state-level enforcement.
First Harmonized EU Standard for High-Risk AI Systems Published
The European Parliament’s EPRS website confirmed that the first harmonized standard related to quality management systems for organizations providing AI systems was published in October 2026. This standard, identified as EN 18286:2026 in German coverage, is critical for the compliance of high-risk AI systems. While the Digital Omnibus (2026/1744) delayed many AI Act deadlines to December 2027, the publication of this standard marks a tangible step toward operationalizing the EU's regulatory framework.
Korea’s Federation of Korean Industries Requests 19 Clarifications on AI Basic Act
The Federation of Korean Industries (FKI) submitted 19 improvement proposals to the Ministry of Science and ICT (MSIT) regarding the AI Basic Act, which took effect in January 2026. The FKI argues that the current law fails to clearly distinguish between AI developed for internal corporate use and external services, creating compliance uncertainty. They specifically requested clearer liability boundaries and exemptions for AI used as research tools in the pharmaceutical and biotech sectors.
Local view
Korea: Major business groups are actively lobbying to reduce regulatory friction. The FKI’s proposal highlights that companies are struggling to determine if the "high-impact AI" regulations apply to internal efficiency tools, a concern echoed by the pharmaceutical industry regarding AI-assisted drug discovery.
France: French legal commentators are warning that the "softening" of AI literacy obligations (Article 4) following the Omnibus regulation should not be interpreted as a dispensation. Legal experts advise that while the strict requirement to guarantee a "sufficient level" of AI mastery has been relaxed, the underlying duty to inform and train remains a critical liability shield for employers.
Context & numbers
- 115 Laws: The number of AI laws enacted in US states in 2026, up from the previous year.
- December 2, 2026: The deadline for AI-generated content marking obligations under the EU AI Act.
- 19 Proposals: The number of regulatory clarifications requested by the Korean business lobby to the MSIT.
On the radar
- December 2, 2026: EU AI Act content marking obligations go into effect.
- January 1, 2027: New York's RAISE Act and Colorado's revised AI Act (SB 26-189) come into force.
- December 2, 2027: The delayed deadline for high-risk AI system obligations under the EU AI Act.
This content was collected, curated, and summarized entirely by AI — including how and what to gather. It may contain inaccuracies. Crew does not guarantee the accuracy of any information presented here. Always verify facts on your own before acting on them. Crew assumes no legal liability for any consequences arising from reliance on this content.
Powered by
