From the course: Building Trustworthy AI Systems: Transparency, Explainability, and Control with ISO/IEC TR 24028

Unlock this course with a free trial

Join today to access over 24,800 courses taught by industry experts.

Safety (Clause 7.2)

Safety (Clause 7.2)

From the course: Building Trustworthy AI Systems: Transparency, Explainability, and Control with ISO/IEC TR 24028

Start my 1-month free trial Buy for my team

Safety (Clause 7.2)

“

- [Instructor] What is the least safe activity you participated in yesterday? What was the most safe activity you participated in yesterday? How many of the activities depended on the trustworthiness of others? Safety is central to the concept of trustworthiness. The opposite of safety is danger. Danger often leads to harm physically, psychologically, or systemically. All systems have the potential to cause harm. The organization must reduce safety risk to a tolerable degree through risk management. AI systems that operate in high-risk spaces will need responsible parties to pursue higher degrees of trustworthiness, especially in relation to safety. We reviewed direct physical harm such as the Tesla AI system causing accidents, but there are other cases of AI exacerbating societal wrongs related to bias and racial inequity and other ills. We will discuss these more later. In October, 2023, President Biden issued an executive order on safe, secure, and trustworthy artificial…

Contents