Standardizing measurements of autonomy in the artificially intelligent

  • Authors:
  • Amy R. Hudson;Larry H. Reeker

  • Affiliations:
  • University of Maryland College Park, Gaithersburg, MD;National Institute of Standards and Technology, Gaithersburg, MD

  • Venue:
  • PerMIS '07 Proceedings of the 2007 Workshop on Performance Metrics for Intelligent Systems
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

The amount of control that an intelligent system has over their actions, whether they are able to act independently from their creator, plays a major factor in describing systems and in distinguishing them from each other. Different levels of autonomy reflect the different abilities of the machines as well as where and how they can play a part in our daily lives. We may begin to comprehend these abilities and possible applications into human society once we can classify the levels of autonomy. The goal of this project is to set a framework for establishing a standard of autonomy in the scientific community by examining past and current methods of measurement as well as exploring different levels of autonomy's ethics and implications. Once this framework is made available to the scientific community, more tests and experiments will be conducted to refine and further ingrain the framework so that classifications of artificial intelligent agents are available universally. If we are to continue improving upon our machines, developing them to be more adept at communicating and accomplishing tasks, then a set of standards must be established for the safety and convenience of mankind.