The EU regulates AI through a binding, risk-based law; the US approach described in current federal policy relies on executive orders alongside agency and sector-specific measures, amid a dispute over state laws; and China’s cited framework consists of administrative rules for particular kinds of internet services. The difference is not simply how strict each system is: it is also what kind of instrument creates the rules, which uses or services they address, and how their timelines work.
How the three approaches differ
| Jurisdiction | Legal form in the cited framework | Organizing approach |
|---|---|---|
| United States | Executive orders, federal agency policy and sector-specific instruments; state laws are part of an active federal-state dispute. | Federal policy emphasizes US AI leadership and a national framework, while the cited executive orders direct actions rather than themselves creating a comprehensive AI statute. |
| China | Cyberspace Administration of China (CAC) administrative provisions for defined internet service areas. | Service-specific rules covering algorithm recommendations, deep synthesis and generative AI services. |
| European Union | Regulation (EU) 2024/1689, a harmonized, binding regulation. | Risk-based duties for specified AI uses, ranging from prohibited practices to high-risk requirements and transparency obligations. |
This comparison concerns the instruments identified here, not every law that may affect AI in each jurisdiction. It is a policy overview, not jurisdiction-specific legal advice.
![AI Law and Policy: [Connected eBook]](https://m.media-amazon.com/images/I/41Y7jOay2eL._SL160_.jpg)
What the US federal approach means
The sources reviewed do not establish a single, comprehensive federal AI statute comparable to the EU AI Act. The current federal direction is expressed through executive orders, agency policy and sectoral law; those instruments should not be treated as interchangeable with an act of Congress.
Executive Order 14179
Signed on 23 January 2025, Executive Order 14179 states that US policy is to “sustain and enhance America’s global AI dominance” to promote human flourishing, economic competitiveness and national security. It directed development of an AI Action Plan and review of actions taken under the previously issued Executive Order 14110. These are executive-branch policy directions, not a comprehensive statutory code for AI.
Executive Order 14365 and state laws
Signed on 11 December 2025, Executive Order 14365 directs the creation of an AI Litigation Task Force, evaluation and challenges to certain state AI laws, steps concerning certain federal funding and reporting standards, and preparation of a legislative recommendation for a uniform framework. The order also contemplates exceptions related to child safety, compute and data-center infrastructure, and state procurement or use.
Those directives do not mean that every state AI law has been preempted, that a court has invalidated the laws at issue, or that Congress has enacted the proposed uniform framework. The federal-state boundary remains a live policy and legal issue; other federal rules and litigation have their own timelines.
Rank #2
What China’s cited AI rules cover
The CAC’s rules provide a service-specific picture rather than one general AI statute in the materials considered here. The relevant primary-source starting points are three Chinese-language provisions:
- Algorithm recommendations: Provisions on the Administration of Algorithmic Recommendations in Internet Information Services, dated 4 January 2022.
- Deep synthesis: Provisions on the Administration of Deep Synthesis Internet Information Services, dated 11 December 2022.
- Generative AI services: Interim Measures for the Management of Generative Artificial Intelligence Services, issued 13 July 2023.
The documents identify the service areas addressed, but their titles alone do not establish the precise coverage thresholds, duties, exemptions or penalties. Those details should be checked against the legal text and an authoritative translation before being applied to a particular service or organization. Each instrument also has its own issuance and implementation dates; do not assume one shared effective date.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →Rank #3
How the EU AI Act sorts uses and obligations
Regulation (EU) 2024/1689 sets a common EU framework for AI developers and deployers. The European Commission describes four risk levels: unacceptable, high, transparency or limited, and minimal or no risk. The level matters: the Act does not treat every AI system as high risk.
Prohibited practices and higher-risk systems
Among the practices the Commission lists as prohibited are social scoring, some biometric identification and categorization uses, and emotion recognition in workplaces or educational institutions. For high-risk systems, the Act provides for duties that include risk assessment and mitigation, data quality, logging, documentation, information for deployers, human oversight, and accuracy, robustness and cybersecurity.
Staged application and the 2026 changes
The Act’s requirements take effect in stages. Prohibitions and AI literacy obligations began applying on 2 February 2025; governance and general-purpose AI provider obligations began applying on 2 August 2025. The general application date and transparency duties arrived on 2 August 2026.
The European Commission reports that the AI Omnibus entered into force on 27 July 2026. Under the Commission’s current timeline, high-risk requirements for Annex III use cases apply from 2 December 2027, while those for high-risk AI systems embedded in regulated products apply from 2 August 2028. The applicable rule can therefore depend both on the system’s classification and on which category it falls into. For a specific compliance decision, consult the current consolidated legislation and applicable implementation guidance.
Free tools Windows power users keep installed
One-click scans. No signup required.
How to read the comparison in practice
- Start with the instrument. In the EU, begin with the AI Act and the system or use category it addresses. For US federal policy, distinguish an executive order from enacted legislation and from agency or sector-specific rules. For China, identify which service-specific CAC provision may apply.
- Check the relevant date. EU obligations are staged, the US orders have their own dates and do not settle the timing of other rules or litigation, and the Chinese provisions have separate timelines.
- Do not infer scope from a label. Calling a system “AI” does not by itself establish that it is high risk under the EU framework or that a particular CAC measure applies. The applicable text and facts about the system or service matter.
For primary references, consult the European Commission’s AI Act overview and the texts of Executive Orders 14179 and 14365. The CAC hosts the three Chinese-language provisions named above. Legal requirements and implementation guidance can change, so verify the current text for the relevant jurisdiction and use case.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
