Core Viewpoint - The article discusses the contrasting positions of major tech companies regarding the EU's General AI Act and the voluntary AI Code of Conduct, highlighting the implications for innovation and regulatory compliance in the AI sector [1][4][7]. Group 1: EU AI Regulation and Industry Response - The EU's General AI Act will come into effect on August 2, 2025, with key provisions affecting general AI models [1][8]. - Over 45 influential European companies have called for a two-year pause on the strictest requirements of the AI Act, expressing concerns that current regulations may hinder innovation and competitiveness [8][9]. - Meta is the only major tech company to explicitly refuse to sign the AI Code of Conduct, citing legal uncertainties and additional requirements beyond the AI Act [6][9]. Group 2: Company Positions on the AI Code of Conduct - French AI company Mistral was the first to commit to the AI Code of Conduct, followed by OpenAI, which emphasized the importance of simplifying regulatory requirements [5][6]. - Anthropic also announced plans to sign the Code, highlighting its commitment to transparency and accountability in AI development [5][6]. - Microsoft indicated a strong likelihood of signing the Code, aiming to play a rational role in AI governance amid geopolitical tensions [6]. Group 3: Regulatory Challenges and Industry Concerns - The EU's AI Act employs a risk-based regulatory model, imposing strict limitations on high-risk AI applications while requiring transparency and accountability from other AI systems [8][10]. - Critics argue that the EU's regulatory approach may stifle innovation, with many small and medium enterprises facing burdensome administrative processes [10]. - The EU is working on proposals to streamline conflicting legal requirements and reduce compliance burdens, aiming to create a more favorable environment for innovation in AI [10].