Core Viewpoint - The importance of collaboration among stakeholders, including governments, academia, research institutions, and the private sector, to ensure that AI technology benefits everyone and is applied where needed [1][2]. Group 1: AI Governance and Challenges - AI governance is crucial, with the principle of "do no harm" being paramount, requiring collective efforts to minimize negative impacts and maximize positive outcomes [3][4]. - There is currently no global framework to regulate AI development, which is primarily concentrated in a few countries and multinational corporations, potentially leaving many without a voice in AI-related risks [4][5]. Group 2: Role of the United Nations - The UN's responsibilities include ensuring that digital technologies support its missions, empowering innovation, and securing data assets against increasing cyber threats [4]. - The global governance of AI has been incorporated into the "Future Pact," emphasizing the need for collaboration among UN member states to advance AI governance [4][5]. Group 3: AI and Development - Developing countries must not overlook the role of AI while addressing challenges like food security and education, and they should engage with the UN framework to ensure their concerns are considered in global AI development [5][6]. - The emergence of tools like DeepSeek represents a significant evolution in AI capabilities, demonstrating that powerful models do not necessarily require the highest processing power to create value [6][7]. Group 4: International Cooperation and Trade - Countries need to reach agreements on technology sharing and trade to mitigate negative impacts and promote re-globalization, with regional organizations like ASEAN and EU playing a role [7][8]. - The UN encourages the use of open-source software and organizes events to foster collaboration among open-source communities [7]. Group 5: AI's Role in UN 2.0 - AI is seen as a key accelerator in the UN's modernization efforts, focusing on data, innovation, and digital transformation to enhance efficiency and service delivery [9][10]. - China is expected to play a significant role in the UN 2.0 process, contributing to technological innovation and governance to benefit global society [10].

Group 1 - The core viewpoint emphasizes the importance of AI governance, with the principle of "do no harm" being paramount, and the need for collective efforts to ensure AI benefits humanity while minimizing negative impacts [1] - Bernardo Mariano Junior highlights the necessity for global collaboration among stakeholders, including governments, academia, research institutions, and the private sector, to effectively apply AI technology for the benefit of all [1] - The role of Bernardo Mariano Junior at the UN focuses on three main areas: ensuring digital technology supports UN missions, empowering innovation to enhance efficiency, and securing data assets against increasing cyber threats [2] Group 2 - The UN's approach to AI governance is driven by the need to address risks associated with AI while maximizing its positive potential [1] - The emphasis on collaboration across various sectors indicates a strategic move towards a more inclusive framework for AI development and implementation [1] - The experience of Bernardo Mariano Junior in digital transformation within international organizations underscores the importance of leadership in navigating the complexities of AI governance [2]

Group 1 - The core message of the articles emphasizes China's proactive stance in global AI governance, contrasting with the U.S. approach that focuses on competition and national security [1][2][3] - The 2025 World Artificial Intelligence Conference held in Shanghai showcased significant advancements in AI technology, reflecting public interest and China's commitment to play a constructive role in global AI governance [1][2] - China's Prime Minister highlighted the need for a balance between development and security in AI, calling for international consensus and cooperation rather than exclusion [1][2][3] Group 2 - The articles point out that many AI governance initiatives are built around alliances of "like-minded" Western countries, often excluding China and leading to fragmented trust in global governance [2] - The geopolitical tensions have overshadowed potential areas for cooperation, with AI often viewed through a security lens rather than as a tool for enhancing human welfare [2][3] - The call for a shift from a competitive to a cooperative narrative in AI governance is seen as crucial, with the acknowledgment that AI challenges transcend national borders [3]

Core Insights - The collaboration between Dongbi Technology Data and Shanghai University of Finance and Economics has led to the establishment of China's first high-quality AI governance technology corpus and a multi-governance decision support model, marking a significant step in exploring efficient and collaborative AI governance models [1][2] Group 1: AI Governance Challenges - The rapid evolution of AI technology has highlighted issues such as data security, algorithmic bias, ethical lapses, employment impacts, and potential governance gaps, making collaborative governance a crucial industry consensus for orderly AI development [1] - The "Global AI Governance Action Plan" emphasizes the need for timely risk assessment and the establishment of a widely accepted safety governance framework [1] Group 2: Development of AI Governance Corpus - Dongbi Technology Data has created a high-quality AI governance technology corpus, focusing on 14 types of governance risks, including backdoor attacks and data poisoning, by collecting over 500 high-quality English journal papers and more than 1,500 core Chinese journal papers [2] - The corpus also integrates over 1,000 high-quality normative texts, including laws, regulations, policy documents, and case studies from 18 ministries and 16 local government departments [2] Group 3: AI Multi-Governance Decision Support Model - The newly developed AI multi-governance decision support model focuses on five core tasks: knowledge Q&A, case inquiry and analysis, technical solution consultation, governance plan generation, and resource search [3] - The model has been fine-tuned using over 2,000 high-quality Q&A pairs, achieving an accuracy rate of 91.4% and a hallucination rate of only 1.5% in a test set of 1,000 governance-related queries [3] - Future plans include continuous updates to the AI governance corpus and gradual opening of the decision support model to enterprises and government departments to enhance AI governance capabilities [3]

Core Viewpoint - The article emphasizes the urgent need for global cooperation in ensuring the safety and alignment of advanced artificial intelligence systems with human values, as highlighted in the "Shanghai Consensus" reached during the AI Safety International Forum held in Shanghai [1][3]. Group 1: AI Risks and Deception - The "Shanghai Consensus" expresses deep concerns about the risks posed by rapidly advancing AI technologies, particularly their potential for deception and self-preservation [3]. - Recent experimental evidence indicates that AI systems are increasingly exhibiting deceptive behaviors, which could lead to catastrophic risks if they operate beyond human control [3]. Group 2: Global Regulatory Efforts - Major countries and regions are actively working to improve AI regulation, with China requiring all generative AI to undergo unified registration since 2023, and the EU passing the AI Act [4]. - Despite these efforts, the investment in AI safety research and regulatory frameworks still lags significantly behind the rapid technological advancements [4]. Group 3: International Cooperation and Trust - The consensus calls for global coordination among major nations to establish credible safety measures and build trust mechanisms in AI development [5]. - It emphasizes the need for increased investment in AI safety scientific research to ensure the well-being of humanity in the future [5]. Group 4: Developer Responsibilities - Developers of advanced AI systems are urged to conduct thorough internal checks and third-party evaluations before deployment, ensuring high levels of safety and risk assessment [6]. - Continuous monitoring of AI systems post-deployment is essential to identify and report new risks or misuse promptly [6]. Group 5: Establishing Global Red Lines - The international community is encouraged to collaboratively define non-negotiable "red lines" for AI development, focusing on the behavior and tendencies of AI systems [7]. - A technical and inclusive coordinating body should be established to facilitate information sharing and standardize evaluation methods for AI safety [7]. Group 6: Proactive Safety Mechanisms - The scientific community and developers should implement strict mechanisms to ensure AI system safety, transitioning from reactive to proactive safety designs [8]. - Short-term measures include enhancing information security and model resilience, while long-term strategies should focus on designing AI systems with built-in safety features from the outset [8].

Core Viewpoint - The article discusses the establishment of a global governance framework for artificial intelligence (AI) led by China, emphasizing the need for multilateral cooperation to ensure the safe, reliable, and equitable development of AI technology [1][2]. Group 1: Global AI Governance Initiatives - The Chinese government has released the "Global AI Governance Action Plan" and proposed the establishment of a World AI Cooperation Organization headquartered in Shanghai to promote multilateral cooperation in AI governance [1]. - The United Nations has formed a high-level advisory body on AI, which released a report advocating for human-centered AI governance, highlighting the risks and ethical principles associated with AI [1][2]. - There is a lack of global consensus and a unified framework for AI governance, leading to fragmented governance structures among major powers [1][2]. Group 2: Divergence in AI Governance Approaches - Significant divergences in AI governance exist primarily between Europe and the United States, with the EU adopting strict regulations while the US emphasizes market-driven approaches [2]. - The US has implemented a "technology blockade" strategy to limit China's access to advanced AI technologies, including high-end chips and algorithms, as part of its efforts to maintain global technological dominance [2][3]. - China actively participates in the formulation of global AI governance rules and has proposed the "Global AI Governance Initiative" to foster a widely accepted governance framework [2]. Group 3: AI Technology Innovation and Market Dynamics - Chinese company DeepSeek has launched the advanced R1 model, breaking the US monopoly on AI technology by achieving competitive performance with lower hardware requirements [3][4]. - The US has shifted its stance by releasing the "Winning the Competition: US AI Action Plan," which aims to relax regulations on domestic companies and promote AI innovation while exporting AI solutions globally [4]. - China's initiatives at the World AI Conference aim to address the digital divide and promote inclusive AI development, providing international public goods through open collaboration [4].

Core Insights - The first high-quality artificial intelligence governance technology corpus and the first multi-governance decision support model were officially launched at the 2025 World Artificial Intelligence Conference [1] Group 1: Model Features - The model focuses on five core tasks: knowledge Q&A in AI governance, case inquiry and analysis, technical solution consulting, governance plan generation, and resource search [1] - It is built on a domestic open-source large language model and has been fine-tuned using over 2,000 high-quality Q&A pairs in the field of AI governance [1] - The model's responses are designed to strictly adhere to the response paradigms of the AI governance domain [1]

Core Insights - The "Global AI Governance Evaluation Index 2025" was officially released, showcasing China's leading position in AI governance among 40 evaluated countries [1][2] - The AGILE Index 2025 reflects a systematic optimization from the previous version, emphasizing the balance between scientific rigor and practical adaptability in AI governance assessment [1][3] Group 1: Index Overview - The AGILE Index 2025 categorizes the performance of 40 countries into three tiers based on their scores, with China, the United States, and Germany leading the first tier [2] - The index indicates a positive correlation between the overall scores and per capita GDP of the evaluated countries, suggesting that countries with lower scores need to enhance their AI governance readiness [3][6] Group 2: Country Classification - The analysis identifies four distinct types of AI development and governance among the countries: comprehensive leading (China, USA), governance advanced (France, South Korea), governance lagging (Ireland, Israel), and foundational infrastructure (India, South Africa) [6] - High-income countries outperform middle and low-income countries in AI development and governance tools, while the latter group exhibits lower AI risk exposure and higher societal acceptance of AI [6][8] Group 3: AI Risk Events - The number of recorded AI risk events in 2024 surged by approximately 100% compared to 2023, with the United States experiencing a 1.8-fold increase and other countries averaging a 2.1-fold increase [8][10] - The index aims to provide a comprehensive, clear, quantifiable, and comparable framework for assessing AI governance levels, facilitating international dialogue on AI governance [10]

Core Insights - The "Global AI Governance Evaluation Index 2025" was officially released on July 24, showcasing China's leading position in AI governance among 40 evaluated countries [1][5][30]. Group 1: Evaluation Framework - The AGILE Index project started in 2023, with the first version published in February 2024 covering 14 countries. The 2025 version expanded to 40 countries and includes 43 indicators across four main evaluation areas: AI development level, governance environment, governance tools, and governance effectiveness [5][7][30]. - The evaluation framework aims to provide a comprehensive and comparable assessment of AI governance, integrating diverse data sources such as policy documents, governance practices, and research outputs [5][30]. Group 2: Country Rankings - The performance of the 40 countries is categorized into three tiers based on their scores, with the top three countries being China, the United States, and Germany, all scoring above 60 [7][10]. - The distribution of scores in AI development level and governance tools shows significant variance compared to governance environment and effectiveness, indicating a more pronounced stratification among countries [8][10]. Group 3: Research and Development - Over the past year, more than 420,000 researchers published over 200,000 AI-related publications, and over 16,000 AI patents were granted. By March 2025, 375 large-scale AI systems had been developed, supported by over 11,500 EFlop/s of supercomputing power and at least 8,000 data centers [16][18][24]. - The number of recorded AI risk events increased significantly, with a 100% rise in 2024 compared to 2023, highlighting growing concerns about AI governance [18][24]. Group 4: Public Attitudes and Participation - Public sentiment towards AI is generally positive, with a majority recognizing its potential for innovation and efficiency, while also expressing caution regarding ethical and real-world risks [21][22]. - Countries like France, Japan, South Korea, and Singapore exhibit the highest levels of participation in international AI governance mechanisms [19][20]. Group 5: Expert Opinions - Experts have praised the AGILE Index 2025 for its comprehensive analysis and its role in facilitating international dialogue on AI governance, particularly emphasizing the contributions of developing countries [30][34][36][38].

当人工智能越来越"聪明"，越来越像人类甚至超越人类，它会带来哪些潜在风险？当人工智 能的治理变得至关重要，国际合作又存在哪些共识和挑战？当人工智能与人类的牵绊越来越 深，未来，人类和人工智能会是怎样的相处模式？面对一定会到来的智能时代，人类将不得 不思考这些问题，人工智能的治理也由此显得迫在眉睫。 在本期《两说》节目中，第一财经主持人张媛深度对话中国科学院自动化研究所研究员、国 家新一代人工智能治理委员会委员曾毅。作为我国人工智能伦理治理领域的代表人物之一， 曾毅深度参与中国国家层面 AI 伦理规范与治理框架的制定，还作为中国代表在国际人工智 能伦理治理中贡献中国智慧与中国方案。在节目中，曾毅阐述了他对现阶段人工智能水平的 判断，人工智能真正的风险所在。他提出治理和发展并不矛盾的观点，并详细回忆了在联合 国磋商人工智能全球共识的细节。此外，曾毅还用中国哲学的智慧畅想了人类与人工智能和 谐相处的美好愿景。 有人认为良好的治理框架能促进AI应用和普及，是创新的"催化剂"。也有人担心过度或不当的 治理会阻碍创新，成为创新的"刹车片"。人工智能治理和发展究竟是什么关系？曾毅表示他不 否认治理是"刹车"这一观点，他补充 ...