The EU AI Act takes full effect

The European Union’s AI Act is no longer a proposal on paper; it is the law of the land. After entering into force in August 2024, the regulation becomes fully applicable on August 2, 2026. This two-year transition period allowed companies to adapt their systems, but the clock is now ticking for full compliance across the single market.

The EU AI Act entered into force in August 2024 and becomes fully applicable on August 2, 2026.

The core of the Act is its risk-based framework. It categorizes AI systems into four distinct levels: minimal risk, limited risk, high risk, and unacceptable risk. This structure determines the compliance burden placed on developers and deployers. Systems deemed to pose an unacceptable threat to safety or fundamental rights are banned outright. This includes social scoring by governments and real-time remote biometric identification in public spaces for law enforcement, with narrow exceptions for targeted searches of victims or potential perpetrators.

High-risk AI systems face the most stringent requirements. These include AI used in critical infrastructure, education, employment, and essential private and public services. To market these systems in the EU, providers must ensure data governance, technical documentation, transparency, human oversight, and robust cybersecurity. Non-compliance can result in fines of up to 7% of global annual turnover or €35 million, whichever is higher.

Limited risk AI, such as chatbots and deepfakes, faces transparency obligations. Users must be informed when they are interacting with an AI system. For deepfakes, the content must be clearly labeled as artificially generated or manipulated. This allows individuals to make informed decisions about the content they consume.

The vast majority of AI applications, including spam filters or video games, fall into the minimal risk category. These systems are not subject to specific regulatory requirements under the Act. This approach aims to foster innovation while protecting citizens from the most significant harms. Companies operating globally should review their AI portfolios to identify which systems fall into the high-risk or limited-risk categories before the August 2026 deadline.

For the latest official updates and legal text, refer to the European Commission’s AI Act page. The regulatory landscape is complex, and legal counsel is recommended for specific compliance strategies.

US states fill the federal gap

With no comprehensive federal AI law in place, compliance has shifted to the state level. Four states—Colorado, California, Texas, and Illinois—have enacted binding regulations that create a fragmented but mandatory landscape for any organization deploying automated systems. Rather than a single national standard, businesses now navigate a patchwork of overlapping requirements.

Colorado’s Artificial Intelligence Act, which took effect in February 2026, was the first of its kind. It mandates impact assessments, transparency disclosures to consumers, and documentation of AI decision-making processes. The law requires developers to implement reasonable safeguards to protect against bias and to inform users when they are interacting with an AI system. This foundational approach has influenced subsequent legislation in other states.

California has followed with its own aggressive regulatory framework, focusing heavily on algorithmic discrimination and consumer rights. The state’s laws often impose stricter documentation requirements and higher penalties for non-compliance, particularly regarding high-risk AI applications in housing, employment, and credit. Companies operating in California must ensure their models are auditable and that their risk management practices meet specific state-defined thresholds.

Texas and Illinois have also introduced distinct regulations. Texas has focused on transparency in hiring algorithms and ensuring that automated systems do not discriminate in employment decisions. Illinois, known for its biometric data laws, has expanded its regulatory reach to include AI-generated content disclosures and stricter controls on automated decision-making in sensitive areas. These state-level laws collectively form the primary compliance burden for US-based AI deployments.

The result is a complex compliance environment where companies must map their AI use cases against multiple state statutes. What is legal in one jurisdiction may be prohibited in another, requiring organizations to adopt a "highest common denominator" approach to risk management. This fragmentation increases legal costs and operational complexity, particularly for national platforms.

Federal executive orders guide policy

The United States has chosen a distinct path from the European Union’s comprehensive AI Act. Instead of broad legislative mandates, federal policy relies on targeted executive orders to balance national security with innovation. This approach allows the White House to adjust regulations quickly in response to rapid technological shifts without waiting for congressional approval.

Executive Order 14409, issued in June 2026, serves as the cornerstone of this strategy. Titled "Promoting Advanced Artificial Intelligence Innovation and Security," the order directs federal agencies to establish rigorous safety standards for advanced models while encouraging private-sector development. It emphasizes that security and innovation are not mutually exclusive but rather complementary goals that require coordinated federal oversight.

The order focuses heavily on data integrity and model transparency. It mandates that developers of frontier AI systems share safety test results with the government. This requirement aims to prevent the deployment of models that pose significant risks to national security or public safety. By creating a formal channel for information sharing, the federal government seeks to mitigate potential harms without stifling the competitive edge of US tech firms.

This executive framework complements existing enforcement actions by agencies like the Federal Trade Commission (FTC). While the executive orders set the strategic direction, the FTC continues to use Section 5 of the FTC Act to police deceptive AI practices. This dual-layer approach ensures that both national security concerns and consumer protection issues are addressed through different regulatory mechanisms.

The reliance on executive action reflects a pragmatic response to the pace of AI development. It allows the US to maintain a flexible regulatory environment that can adapt to emerging threats and opportunities. However, this method also lacks the permanent legislative backing that would provide long-term certainty for businesses operating in the AI space.

Global frameworks shape compliance

The United States and European Union do not operate in a vacuum. Through 2026, distinct regulatory models in the US, EU, China, and the UK enforce different approaches to AI governance, creating a complex web for multinational organizations. Compliance is no longer a regional checklist but a global alignment challenge.

South Korea offers a distinct model with its AI Basic Act, which took effect on January 22, 2026. This framework combines industrial promotion with baseline obligations for trust, safety, transparency, and user protection. Unlike the EU’s risk-based tiering, the Korean approach integrates development incentives directly with safety mandates, signaling a shift toward state-led AI stewardship.

This fragmentation is accelerating. At least 72 countries have proposed over 1,000 AI-related policy initiatives, according to industry tracking. For businesses, this means that adhering solely to US state laws or the EU Act is insufficient. A robust compliance strategy must account for these divergent global frameworks, particularly in regions like East Asia, where regulatory structures are evolving rapidly alongside technological adoption.

Common questions on AI policy

AI regulation is shifting from abstract principles to enforceable law. In 2026, the landscape is defined by a patchwork of state laws in the US, active federal enforcement, and new frameworks in countries like South Korea. Below are answers to the most frequent queries about how these rules impact jobs, compliance, and policy definitions.

What is the AI policy 2026?

The "AI policy 2026" generally refers to the implementation phase of major regulatory frameworks that took effect or entered enforcement in 2026. In the EU, the AI Act is now fully operational, mandating risk-based compliance. In the US, while there is no single federal law, states like California, Colorado, and Texas have active rules. These laws typically require impact assessments, transparency disclosures to consumers, and documentation of AI decision-making processes [src-serp-7].

Is AI going to be regulated in the US?

Yes, but through a combination of state laws and federal enforcement rather than a comprehensive federal statute. The Federal Trade Commission (FTC) is actively using Section 5 of the FTC Act to fine firms for unfair or deceptive practices involving AI. This includes misleading claims about AI capabilities or harmful outputs. Simultaneously, state-level regulations in Colorado, California, Texas, and Illinois create a complex compliance web for companies operating across state lines [src-serp-4].

What 3 jobs will not be replaced by AI?

While AI automates routine tasks, jobs requiring high levels of human empathy, complex physical dexterity, and strategic ethical judgment remain resistant to automation. Roles in healthcare (such as nursing and therapy), skilled trades (like plumbing and electrical work), and senior leadership (requiring nuanced negotiation and accountability) are widely considered secure. These positions rely on interpersonal trust and physical adaptability that current AI systems cannot replicate.

What is the AI Basic Act 2026?

The AI Basic Act is South Korea’s national framework that took effect on January 22, 2026. It combines AI industrial promotion with baseline obligations for trust, safety, transparency, and user protection. Unlike the EU’s risk-based approach, the Basic Act focuses on establishing a foundational legal structure to foster innovation while ensuring public safety, serving as a model for other Asian nations [src-serp-7].