Group 1 - The core viewpoint of the article emphasizes the need for global consensus on AI governance to maximize benefits and minimize risks, highlighting China's initiative to lead this effort [1][6]. - China is proposing the establishment of a global AI coordination body, the World AI Cooperation Organization, which aligns with the interests of all nations and encourages government participation [3][6]. - The article notes that while AI models have significant potential for scientific and economic advancement, they also pose risks such as exacerbating inequality and spreading misinformation, which have not been adequately addressed in the current competitive landscape [5][6]. Group 2 - The article points out that the United States lacks a unified regulatory body for AI, relying instead on fragmented state legislation and self-regulation by companies, which has resulted in low safety ratings for major tech firms [5][6]. - In contrast, China is actively integrating AI across various sectors and has implemented regulations requiring safety assessments for AI models, including embedding identifiable markers in generated content to prevent fraud [6][8]. - The global governance of AI is seen as a necessity, with existing frameworks like the OECD's AI Principles and the EU's AI Framework Convention being criticized for their lack of enforceability and effectiveness [8].

Core Viewpoint - The European Union (EU) is planning to relax certain digital regulatory frameworks, including the AI Act, which was initially designed with strict regulations. This shift raises questions about the reasons behind the change and its implications for the AI industry in Europe and globally [3][4][5]. Group 1: Reasons for Initial Strict Regulation - The EU's strict regulatory stance was influenced by its economic structure, which is dominated by small and medium-sized enterprises (SMEs). In 2022, SMEs accounted for 99.8% of non-financial enterprises in the EU, employing 64.4% of the workforce and contributing 51.8% of economic value added. This demographic necessitated clear rules to protect against potential risks associated with emerging technologies [6]. - Politically, strict regulation was seen as a means to maintain digital sovereignty, as Europe has historically lagged behind the US and China in key technological domains. The EU aimed to use regulations as a tool to influence global competition and embed European values into the future AI governance framework [7][8]. - Culturally, the EU emphasizes ethics and rights, leading to a governance approach that prioritizes risk prevention. This is reflected in the long-standing "precautionary principle" that shapes its regulatory logic, particularly in technology that could impact labor rights and public resources [9][10]. - The EU's complex political structure, comprising 27 member states with diverse priorities, naturally leads to stricter regulations as a means of achieving political consensus [11]. Group 2: Reasons for Regulatory Relaxation - The emergence of tangible benefits from AI technology has shifted the risk-reward balance. As AI capabilities have advanced, the economic returns have become more apparent, prompting the EU to reconsider its initial cautious approach [13][14]. - AI technology has become more governable, with advancements in alignment, explainability, and controllability. This has led to a perception that AI can be managed within a regulatory framework, reducing the need for stringent oversight [15]. - The EU's regulatory logic has shifted from a strict "precautionary principle" to a more balanced "proportionality principle," allowing for regulatory measures only when risks are clearly identified [16]. - Geopolitical pressures have also influenced the EU's regulatory stance, as competition with the US and China has highlighted the risks of falling behind in technological innovation [17][18]. - Internal political dynamics within the EU have shifted, with a growing emphasis on industry competitiveness over strict ethical considerations, leading to a more lenient regulatory approach [19][20]. Group 3: Expected Adjustments to the AI Act - The implementation timeline for the AI Act is expected to be delayed, allowing more time for companies to adapt to the regulations. This includes extending grace periods for compliance with high-risk AI system obligations [21][22]. - Obligations for general AI models are likely to be weakened, with a shift from government-led regulation to industry self-regulation through non-binding codes of practice [23][24]. - Penalty provisions are anticipated to transition towards a "warning first" approach, significantly reducing the severity of fines for non-compliance [25][26]. - Discussions are underway to refine the definition of "high-risk systems" to focus regulatory efforts on genuinely high-risk applications, potentially alleviating unnecessary burdens on businesses [27]. - The concept of "regulatory sandboxes" is gaining traction, allowing for relaxed regulatory conditions to foster innovation while ensuring safety [28]. Group 4: Implications of Regulatory Changes - The adjustments to the AI Act are expected to reignite the AI innovation ecosystem in Europe, creating a more favorable environment for local AI development and reducing compliance burdens on startups [29]. - The global AI competitive landscape may shift, moving from a single regulatory paradigm to a multi-centered approach, with different regions adopting varied governance models [30][31]. - Multinational companies will benefit from increased flexibility in their AI strategies, accelerating the diffusion of AI technologies across different sectors [32][33]. - The EU's regulatory changes may foster a new paradigm of "gentle regulation," promoting a balance between oversight and innovation, which could influence global regulatory practices [34][35].

Group 1 - Over 110 institutions, including ASML, SAP, and Mistral AI, have called on the EU to delay the implementation of new AI regulations, emphasizing the need for a more competitive environment for innovation [1] - The core demand from the business sector focuses on the lack of execution details and the pace of regulatory implementation, as the strictest provisions of the EU AI Act are set to take effect in August [1] - The working group, composed of scholars, developers, and rights groups, is still discussing specific execution guidelines, causing significant delays compared to earlier expectations [1] Group 2 - The controversy centers around the EU's proposed voluntary compliance framework, with tech companies criticizing the stringent requirements for third-party model audits and copyright tracing [2] - The EU AI Act, as the world's first comprehensive AI legislation, establishes a tiered regulatory system, imposing heavy penalties on non-compliant companies, including fines of up to 7% of annual revenue for violations [2] - The high compliance costs and vague guidelines are diminishing the attractiveness of the European AI industry, prompting the EU to seek a new balance between regulatory strength and innovation vitality [2]