Global AI Regulation Efforts Intensify as Governments Grapple with Implementation Challenges

Jul 27, 2026

Governments worldwide are struggling to implement effective artificial intelligence regulations. In the United States, proposed federal legislation like the FRONTIER Act, drafted by Representatives Lori Trahan and Jay Obernolte, threatens to preempt over 1,000 state-level bills, including California's landmark SB53. Meanwhile, game-theoretic research from Cornell University warns that weak safety standards can backfire by shifting safety burdens onto downstream developers. Internationally, India's government is weighing dedicated legislation to combat AI-driven offenses as experts debate whether existing frameworks can handle synthetic media.

Federal preemption of state laws

▪A sweeping federal artificial intelligence law in the United States is expected to conflict with numerous existing state-level artificial intelligence laws.
▪Federal artificial intelligence laws are gradually preempting state-level artificial intelligence laws in the United States.

FRONTIER Act provisions

▪The FRONTIER Act defines Covered Subject Areas to encompass frontier artificial intelligence risk transparency, third-party auditing, independent verification, and incident reporting.
▪Representative Lori Trahan and Representative Jay Obernolte drafted a proposed federal legislative bill known as the FRONTIER Act.
▪Section 9 of the FRONTIER Act prohibits states from adopting or enforcing laws that impose new substantive obligations on artificial intelligence developers within Covered Subject Areas.

State-level AI legislative landscape

▪California Governor Gavin Newsom signed SB53, the Transparency in Frontier Artificial Intelligence Act, into law on September 29, 2025.
▪Estimates suggest there are well over 1,000 artificial intelligence-related bills and laws under consideration at the state level in the United States.
▪The Transparency in Frontier Artificial Intelligence Act requires large frontier developers to publish a frontier artificial intelligence framework describing how they incorporate national and international standards.

Constitutional Supremacy Clause framework

▪The United States Constitution contains the Supremacy Clause in Article VI, Clause 2, which establishes federal laws as the supreme Law of the Land.
▪Under the Supremacy Clause framework, if a federal law and a state-level law conflict, the federal law is designed to prevail and must be enforced by state courts.

Types of legal preemption

▪Field preemption occurs when federal law overwhelmingly dominates a regulatory field, while conflict preemption occurs when federal and state laws impose contradictory requirements on the same party.
▪Implied preemption occurs when the scope of a federal law leaves no space for conflicting state regulation, thereby superseding state law by implication.
▪Conditional preemption occurs when a federal law allows an otherwise preempted state-level law to coexist under specified circumstances, such as a sunset clause.
▪Express preemption occurs when Congress explicitly states within a federal law that the statute supersedes state law.

AI regulation implementation challenges

▪A study in the Proceedings of the National Academy of Sciences suggests that weak or poorly designed safety standards can cause generalist developers to offload safety burdens onto downstream specialists.
▪The Indian government informed the Supreme Court of India that it is considering legislation to address artificial intelligence-driven offenses.
▪Experts in India are divided on whether existing laws like the Information Technology Act are sufficient or if an entirely new legal framework is required to regulate generative artificial intelligence.