《exploring the possibility of using intelligent computers to judge criminal offenders》由会员分享，可在线阅读，更多相关《exploring the possibility of using intelligent computers to judge criminal offenders（9页珍藏版）》请在金锄头文库上搜索。
1、Exploring the Possibility of Using Intelligent Computers to Judge Criminal Offenders Theodore A. Dunn, Jr. Abstract The human race has never encountered a species equal to or greater in intelligence than itself. However, if computer scientists specializing in the field of artificial intelligence (AI
2、) realize their ultimate goal, humans wont be the only intelligent “beings.“ This research explores the possibility that these intelligent beings (computers) might someday be used to replace the functions of the judge and/or jury in a criminal trial. Specific emphasis is placed on two issues: 1) exa
3、mining the history and successes of AI to substantiate the future existence of an intelligent computer; and 2) exploring issues that affect the probability the United States Constitution will be changed in support of the new technology. Introduction This paper examines the possibility that an artifi
4、cial form of intelligence could someday stand in judgement of a human criminal offender. This theoretical concept is not rooted in fantasy but in the fact that significant advances have been made in the fields of computer science and artificial intelligence (AI); advances that scientists hope will l
5、ead to the development of computer systems that have the ability to think, reason, learn, and make decisions much like humans. Berman and Hafner (1989) stated: It is widely viewed that the American legal system is in a state of crisis, plagued by excessive costs, long delays, and inconsistency leadi
6、ng to a growing lack of public confidence. One reason for this is the vast amount of information that must be collected and integrated in order for the legal system to function properly. In many traditional areas of law, evolving legal doctrines have led to uncertainty and increased litigation at a
7、high cost to both individuals and society. (p. 928). When looking for solutions to these problems confronting the American legal system and considering the technological advances experienced in the computer sciences, a reasonable question to ask is: Is it possible that, in the not-so-distant future,
8、 an artificial intelligence might replace the functions of the criminal trial judge and/or jury? Overwhelming implications are associated with the idea that a machine might someday be given judgement over man. Two issues must be examined before attempting to answer the central question. First, the l
9、ikelihood that computer scientists will succeed in producing “intelligent“ computers capable of processing information in much the same way as the human brain must be explored. In doing so, the past and present capabilities of AI are examined, thus giving some insight into the future possibilities o
10、f AI. Second, the process of changing the United States Constitution in support of the new technology must be examined. The constitution guarantees criminal defendants a trial by jury and also sets forth the procedure for changing the constitution. Some perplexing constitutional legal issues need to
11、 be addressed when advocating a change to the jury process. Assuming there is sufficient evidence to support the future existence of truly intelligent computers, overcoming constitutional issues might prove as challenging as the technological advancement of AI. Once these issues have been addressed,
12、 the main questions concerning AIs fitness to stand judgement over man are examined. Artificial Intelligence - Past and Present “AI is the study of cognitive processes using the conceptual frameworks and tools of computer science“ (Rissland, 1990, p. 1957). The study of AI began some 30 years ago wh
13、en scientists from the Rand Corporation proved that computers could do more than calculate; that indeed they could be used to simulate certain aspects of intelligence. Their reasoning was based on the fact that computers are physical symbol systems and that these symbols could theoretically represen
14、t anything, including elements of the real world (Dreyfus, 1993). One of these real world elements was the human process of reasoning. Reasoning, with its roots in formalized beliefs, rules and procedures, seemed to be a function ideally suited to computers. However, after many years of research the
15、 reality of computers with reasoning capabilities, or “thinking computers,“ has somewhat eluded computer scientists and AI practitioners. When pondering the basic question of “Can machines think?“, many AI practitioners now believe there wont be thinking machines until we understand more about how a
16、 human thinks (Newquist, 1993). This becomes a problem because as we learn more about knowledge, intelligence or wisdom, the less we understand them (Angell, 1993). Current technology suggests that AI scientists may be well on their way to realizing their ultimate goal. Neural networks, fuzzy logic architectures, and Bayesian belief networks are currently among the most promising of AI tools. “Neural technology is in essence programming computers with enough intelligence to allow them to lear
