Skip to main content

Safety

A Survey of Artificial Intelligence Challenges: Analyzing the Definitions, Relationships, and Evolutions

Saghiri et al. (2022)

Category
Risk Domain

AI systems acting in conflict with human goals or values, especially the goals of designers or users, or ethical standards. These misaligned behaviors may be introduced by humans during design and development, such as through reward hacking and goal misgeneralisation, or may result from AI using dangerous capabilities such as manipulation, deception, situational awareness to seek power, self-proliferate, or achieve other goals.

The actions of a learning model may easily hurt humans in both explicit and implicit manners...several algorithms based on Asimov’s laws have been proposed that try to judge the output actions of an agent considering the safety of humans

Other risks from Saghiri et al. (2022) (15)