OpenAI Safety Executive Advocates for Responsible AI Development at Bilibili Event
OpenAI safety executive emphasises responsible AI development at Bilibili event. Importance of human involvement in shaping the future of AI highlighted. Chinese researchers at OpenAI praised for their contributions to AI development.
Lilian Weng Li, a research vice-president at the Microsoft-backed company who focusses on AI safety, described AI as a "double-edged sword" during the event. She emphasised the importance of human involvement in navigating the benefits and challenges of AI, urging collaboration to develop intelligent and responsible AI systems.
OpenAI, based in San Francisco and known for innovations such as ChatGPT, does not provide services in China, including Hong Kong and Macau. This decision places the company in a select group of countries subject to US sanctions. In July, OpenAI restricted Chinese users' access to its application programming interface.
Weng, a Peking University and Indiana University Bloomington alumnus, has previously worked for major tech companies including Facebook and Dropbox. She joined OpenAI in 2018 and has since played an important role in furthering AI development. Weng, who was recently promoted to vice-president of research and safety, leads a team focused on implementing practical measures to ensure AI safety, which is a pressing concern in the tech industry and among policymakers.
During her speech, which was delivered in Chinese and broadcast on Bilibili, Weng emphasised the importance of proper AI model training in protecting AI for the benefit of humans. She highlighted the potential risks of data bias and malicious exploitation, emphasising the importance of education, AI security, and alignment research in facilitating AI's smoother growth process.
Weng avoided discussing OpenAI or the US-China tech rivalry, instead focussing on the fundamentals of AI development. She emphasised the importance of robust, diverse, and high-quality data for effective AI training and advocated for a comprehensive approach to developing the framework for AI systems.
Chinese researchers at OpenAI have received praise from mainland media and institutions, demonstrating the growing influence of Chinese talent in the global AI space. According to MacroPolo research, the presence of Chinese scholars in AI research in the United States has increased significantly, with 38% of top-tier AI researchers at US institutions originating from China in 2022, up from 27% in 2019.
While OpenAI does not disclose demographic information about its workforce, a large number of contributors to its most recent model appear to have Chinese surnames. This trend highlights Chinese talent's impact in the field of AI, as well as the recognition they receive domestically and internationally.
OpenAI safety executive emphasises responsible AI development at Bilibili event
Importance of human involvement in shaping the future of AI highlighted
Chinese researchers at OpenAI praised for their contributions to AI development
Source: SCMP