AI Regulation News Today: Global & U.S. Policy Updates

Source: Dev.to

Global AI Regulation Updates

• European Union: The EU Artificial Intelligence Act (Regulation 2024/1689) was adopted in July 2024, entered into force in August 2024, and will have most provisions effective by 2026. It takes a risk‑based approach and can impose fines up to 7 % of global turnover (or €35 million) for non‑compliance.
• China: Introduced its first generative‑AI rules (“Interim Measures”) governing content‑service providers.
• India, Singapore, and others: Rolling out national AI strategies and sector‑specific guidelines.
• International bodies: The UN has urged member states to adopt AI rules for “safe, secure and trustworthy” systems, while the OECD promotes trustworthy AI through its AI Principles.

Key Highlights

• EU AI Act (2024): First comprehensive AI law across the 27‑member bloc; risk‑based framework with substantial penalties.
• China’s Interim Measures: Administrative rules for generative‑AI service providers in China’s digital ecosystem.
• International Frameworks: OECD, G7, and UN resolutions emphasize AI ethics and call for national regulations.

U.S. AI Regulation 2025

The United States still lacks a single federal AI law, relying on a patchwork of existing statutes and agency guidance.

Trump’s 2025 Executive Order – “Removing Barriers to American Leadership in AI”

• Issued in January 2025, it revokes many Biden‑era AI directives, including the October 2023 AI EO (Safe, Secure & Trustworthy AI).
• Directs federal agencies to withdraw policies viewed as obstacles to AI development and to prioritize a permissive, growth‑focused stance.

America’s AI Action Plan (July 2025)

• Outlines more than 90 federal initiatives aimed at securing U.S. AI leadership.
• Emphasizes export of AI technology, infrastructure upgrades, and industry incentives.
• Contrasts with the EU’s risk‑based model and with state laws that focus on bias prevention.

No Comprehensive Federal Law Yet

• Developers continue to operate under existing statutes (consumer protection, aviation, defense, etc.).
• Agencies such as the FTC and FCC are adapting current mandates to address AI‑related issues.
• Congressional proposals largely favor voluntary guidelines or the creation of new agencies rather than stringent mandates, to avoid stifling innovation.

State AI Laws in the United States

State‑level activity is creating a fragmented regulatory landscape.

• Colorado: Passed the nation’s first AI Act in May 2024, effective February 2026. Requires “high‑risk” AI developers and deployers to exercise reasonable care to prevent algorithmic discrimination in areas like hiring, credit, and healthcare.
• California: In 2024, legislators drafted dozens of AI‑related bills covering transparency of AI‑generated content, rights of individuals depicted in AI media, data privacy, and bans on deceptive deepfakes.
• Other States:
  • Utah enacted an AI Policy Act.
  • New York and Illinois are advancing data/biometric laws with AI provisions.
  • Over 45 states considered AI measures in 2024; 31 enacted related laws or resolutions.

Frequently Asked Questions

How is the U.S. handling AI regulation in 2025?

The U.S. has no single AI law. The Trump administration prioritizes innovation, rescinding many Biden‑era directives via the January 2025 EO and launching the July 2025 AI Action Plan with 90+ measures. Congress is debating mostly guideline‑focused bills, while agencies apply existing statutes (e.g., anti‑discrimination rules) to AI.

What is the EU AI Act?

The EU Artificial Intelligence Act, published on 12 July 2024, establishes a risk‑based framework for AI across all member states. High‑risk systems (healthcare, transport, law enforcement, etc.) face strict requirements; prohibited uses (e.g., undetectable manipulative techniques) are banned. The law took effect in August 2024, with most rules enforceable by August 2026, and imposes penalties up to €35 million or 7 % of global turnover.

How many countries have AI regulations?

By early 2025, at least 69 countries have proposed or adopted over 1,000 AI‑related policy initiatives and legal frameworks, ranging from data‑protection adaptations to dedicated AI ethics laws and national strategies.

Which U.S. states have their own AI laws?

• Colorado: Landmark AI Act targeting bias and requiring impact assessments for high‑risk AI.
• California: Numerous bills on AI‑generated content disclosures, safety reporting, and deepfake bans.
• Utah, New York, Illinois, and others: Enacting or considering laws affecting AI use in autonomous vehicles, biometric data, and sector‑specific applications. Over 30 states enacted AI‑related legislation or resolutions in 2024.