On October 18, the forum “Artificial Intelligence and New Trends in Global Development: Sino-Russian Strategic Dialogue” was held at Renmin University of China’s Century Hall, serving as a key sub-forum of the 2025 Tongzhou Global Development Forum. Co-organized by the Huajing School of Artificial Intelligence, the Chongyang Institute for Financial Studies, and the Global Leadership Institute, the event aimed to advance national strategies outlined in China’s “Action Plan on Promoting AI+” by exploring AI’s transformative impact on global development. Over 100 experts, scholars, and students from China and Russia gathered to discuss ethical, governance, and socioeconomic dimensions of AI, advocating for inclusive, sustainable, and human-centered AI development. Renmin University’s Deputy Party Secretary Qing Geliletu emphasized AI’s role in reshaping global power structures and advancing new quality productive forces. He highlighted the university’s leadership in integrating AI with social sciences, citing breakthroughs like the cross-border legal AI model and the open-source Yulan 3.0 large model. He called for joint research platforms, talent development, and inclusive global governance frameworks to strengthen Sino-Russian cooperation in AI. China’s State Council Press Office’s Deputy Director Li Yafang stressed AI as both a driver of economic growth and a tool for addressing global challenges. She urged innovation, strategic dialogue, and practical collaboration between China and Russia, with a focus on advancing AI in international communication and governance. Russian expert Sergei Dmitrievich Bodronov, President of the Russian Free Economy Association and an academician of the Russian Academy of Sciences, described AI as a multi-layered global transformation involving science, engineering, and societal change. He called for new governance models, adaptive systems, and shared values to ensure responsible AI use, emphasizing the strategic synergy between China and Russia in innovation, industry, and security. Professor Liang Zheng from Tsinghua University discussed AI’s impact on both nations, noting China’s rapid progress in large models and applications across healthcare, agriculture, and governance. He pointed to global governance gaps in fairness and capability and advocated for joint efforts within BRICS and SCO frameworks to build a more inclusive, secure AI ecosystem. Professor Zeng Yi from the Chinese Academy of Sciences highlighted a critical imbalance: AI research is concentrated in short-term, high-return sectors like education and medicine, while underfunded areas such as zero hunger and biodiversity remain neglected. He urged embedding ethics and safety as core principles and promoting human-AI “superalignment” to ensure AI serves long-term human well-being. Two roundtables followed. The first, on AI-driven socioeconomic transformation, featured experts discussing how AI can reshape economies, calling for international standards and ethical safeguards. The second, on AI governance and safety, addressed risks like cascading errors—where a 20% error rate per task reduces success to just 32% over five steps—and stressed the need for global cooperation, secure infrastructure, and human-AI collaboration. Meanwhile, Andrej Karpathy, co-founder of OpenAI, expressed skepticism about current AI agents. Speaking on the Dwarkesh Podcast, he criticized their lack of real intelligence, multimodal capability, memory, and continual learning. He believes it will take about a decade to overcome these limitations. While not opposed to AI, Karpathy rejects the vision of fully autonomous agents replacing humans. Instead, he advocates for collaborative coding, where AI assists, explains its actions, and helps humans learn—avoiding the spread of low-quality AI-generated content. He warns that overpromising AI autonomy risks making humans obsolete and undermining trust in AI systems. His view reflects growing concern among experts that current AI capabilities are outpaced by hype, despite long-term optimism.