Formal verification aims to prove whether models satisfy specifications, such as showing a program does what its designer intended. Formal verification is a promising approach that can be used to establish safety, security, and trustworthiness specifications of AI systems. However, to realize the potential societal benefits AI promise, we also need transdisciplinary approaches bridging the gamut from computer science and engineering, the broader sciences, as well as the arts, humanities, social sciences, law, business, and beyond to ensure its development involves all perspectives and voices. My personal AI mission: To develop formal verification methods to help establish and assure the safe, secure, and trustworthy development and use of AI, especially in the context of safety-critical systems such as autonomous cyber-physical systems (CPS).

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Let’s Talk AI with Computer Science Expert Taylor T. Johnson

  • Taylor T. Johnson,
  • Barbara Steffen

摘要

Formal verification aims to prove whether models satisfy specifications, such as showing a program does what its designer intended. Formal verification is a promising approach that can be used to establish safety, security, and trustworthiness specifications of AI systems. However, to realize the potential societal benefits AI promise, we also need transdisciplinary approaches bridging the gamut from computer science and engineering, the broader sciences, as well as the arts, humanities, social sciences, law, business, and beyond to ensure its development involves all perspectives and voices. My personal AI mission: To develop formal verification methods to help establish and assure the safe, secure, and trustworthy development and use of AI, especially in the context of safety-critical systems such as autonomous cyber-physical systems (CPS).