CAREER: How Collaboration and Competition Influence Research in the Emerging Field of Artificial Intelligence

人工智能（AI）是一套颠覆性的技术，可以自动化决策、解决问题和生成内容。人工智能因其对社会的影响而在科学领域中脱颖而出，是研究这种紧张关系的一个有先见之明的案例。人工智能在教育、商业和医学领域带来深刻的社会进步，这是一个充满希望和前景的领域。然而，它也对隐私、司法、安全甚至武装冲突构成重大风险。正如ChatGPT等大语言模型（llm）的公开讨论所表明的那样，人工智能现在既处于科学进步的前沿，也处于国际政治的前沿。该项目研究国际政治如何影响全球人工智能学术研究领域。该研究对比了各国如何引领尖端人工智能研究，并将人工智能作为下一次产业革命的引擎进行投资，以缩小基础研究突破和高端产品开发的差距。本研究探讨了国际政治如何在(1)科学研究的全球性、开放性和普遍性方面，以及(2)国家之间日益增长的竞争性不信任和合作关系的切断之间，为学术研究人员创造了未被充分研究的紧张关系。研究人员进行了三部分、大规模、多方法和相互关联的研究，以展示国际紧张局势如何影响人工智能研究人员、他们的科学以及对全球科学研究的影响。第一个项目部署了社会网络分析和自然语言处理（NLP）技术，通过使用一个名为OpenAlex的数据库，询问是否在其他国家的人工智能学术出版物中出现了不同的国家研究签名，以衡量国际影响力。该项目考察了人工智能研究的国际格局，突出了不同地区和机构之间的差异。它还分析了人工智能研究中国际合作的发展动态，以及不同国家和大学对该领域智力生产的贡献。第二个项目部署了学术人工智能研究人员的调查实验，以确定这种国家影响是公开的还是被低估的。该项目考察了评估不同国家研究的潜在偏见，特别是在人工智能领域，以及隶属关系如何影响对其他国家研究的采用和评估。第三个项目将NLP技术应用于世界各地的研究人员进行的数百次深度半结构化访谈的文本，以显示国家和国际影响如何体现在单个研究人员的合作、工作和职业中。将NLP技术应用于访谈记录可以大规模地揭示主要主题和主题，从而对正在审查的问题有更细致的理解。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Artificial Intelligence (AI) is a set of disruptive technologies that automate decision-making, problem-solving, and content generation. AI stands out among scientific fields for its impact on society and is a prescient case to study this tension. Much hope and promise exist for AI to bring about profound societal improvements in education, commerce, and medicine. However, it also poses significant risks to privacy, jurisprudence, security, and even armed conflicts. As the public discourse on large-language models (LLMs) such as ChatGPT has made clear, AI is now on the frontline of both scientific advancement and international politics. This project investigates how international politics shapes the global field of AI academic research. This research compares how nations lead in cutting-edge AI research and invest in AI as the engine for the next industrial revolution to reduce gaps in basic research breakthroughs and high-end product development. This research explores how international politics creates understudied tensions for academic researchers between (1) the global, open, and universalistic aspects of scientific research and (2) the growing competitive distrust and cutting of collaborative ties between nations. The researcher conducts a three-part, large-scale, multi-method, and interconnected study to demonstrate how international tensions influence AI researchers, their science, and the implications to global scientific research. The first project deploys social network analysis and natural language processing (NLP) techniques to measure international influence by asking whether distinct national signatures of research manifest in academic AI publications of other countries using a database called OpenAlex. The project examines the international landscape of AI research, highlighting variations across regions and institutions. It also analyzes the evolving dynamics of international collaboration in AI research and the contributions of different countries and universities to the field's intellectual production. The second project deploys survey experiments of academic AI researchers to determine how overt or understated this national influence is. The project examines potential bias in evaluating research from different countries, specifically in AI and how affiliation influences the adoption and evaluation of research from other countries. The third project applies NLP techniques to the transcripts of hundreds of in-depth semi-structured interviews with researchers around the world to show how national and international influences manifest in individual researchers’ collaborations, work, and careers. Applying NLP techniques to the interview transcripts can uncover dominant themes and topics at scale, leading to a more nuanced understanding of the issues being examined.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.