主讲人：
陶勇博士是硅谷Cloudstone云石基金的创始合伙人，专注于生物科技、医疗健康、半导体、智能科技和清洁能源等领域的创业投资。他在多家硅谷创新科技公司担任董事职位，覆盖医疗设备、基因检测、半导体和绿色科技。他是世界决策学会（SDP）的研究员，研究方向为基于决策的人工智能治理与立法。在创立云石基金之前，陶勇博士曾建立并领导一家决策管理咨询机构，为移动通信、汽车、清洁能源、科技创新和金融领域的大型企业提供决策咨询服务。他的职业生涯始于硅谷的战略决策集团（SDG），在服务全球500强企业方面积累了丰富的经验，涵盖新药研发、科技、石油、电力和金融等行业。陶勇博士还在多个高等教育及企业培训项目中讲授科学决策和全面风险管理课程，包括国务院发展研究中心-斯坦福联合培训项目和斯坦福职业发展中心等。
讲座简介： The advent of artificial intelligence (AI) offers tremendous benefits but also causes widespread societal unease. The public is deeply concerned about AI because we have not found a robust solution to rein in AI in case of needs. Recent AI legislations are mostly risk-based, they are adequate at identifying and assessing AI risks but fall short in effectively controlling such risks.
This research will arm legislators through the lens of decisions to require AI developers with deep knowledge to build minimum but necessary brakes inside AI to prevent and reduce AI risks, educate users to apply AI wisely, and calm down public fear through a fundamentally new approach to effective solutions to AI risks we urgently need to address.
By examining the core decision elements and processes of both humans and AI and exploring potential human intervention mechanisms based on the solid foundations of decision sciences, we expand the solution space for controlling AI risks to include the crucially important subspaces within the AI system itself. This allows us to develop a systematic framework for comprehensive AI legislation that fundamentally controls AI risks at the decision steps where the danger originates. Together, the controlling mechanisms inside AI, the AI governance infrastructure, and the mitigation of spillover AI risks form the pillars of a sound AI legislation that can effectively and efficiently rein in AI risks to the greatest extent humanly possible.