In this paper we address the question of when a researcher is justified in describing his or her artificial agent as demonstrating ethical decision-making. The paper is motivated by the amount of research being done that attempts to imbue artificial agents with expertise in ethical decision-making. It seems clear that computing systems make decisions, in that they make choices between different options; and there is scholarship in philosophy that addresses the distinction between ethical decision-making and general decision-making. Essentially, the qualitative (...) difference between ethical decisions and general decisions is that ethical decisions must be part of the process of developing ethical expertise within an agent. We use this distinction in examining publicity surrounding a particular experiment in which a simulated robot attempted to safeguard simulated humans from falling into a hole. We conclude that any suggestions that this simulated robot was making ethical decisions were misleading. (shrink)
In their important paper “Autonomous Agents”, Floridi and Sanders use “levels of abstraction” to argue that computers are or may soon be moral agents. In this paper we use the same levels of abstraction to illuminate differences between human moral agents and computers. In their paper, Floridi and Sanders contributed definitions of autonomy, moral accountability and responsibility, but they have not explored deeply some essential questions that need to be answered by computer scientists who design artificial agents. One such question (...) is, “Can an artificial agent that changes its own programming become so autonomous that the original designer is no longer responsible for the behavior of the artificial agent?” To explore this question, we distinguish between LoA1 (the user view) and LoA2 (the designer view) by exploring the concepts of unmodifiable, modifiable and fully modifiable tables that control artificial agents. We demonstrate that an agent with an unmodifiable table, when viewed at LoA2, distinguishes an artificial agent from a human one. This distinction supports our first counter-claim to Floridi and Sanders, namely, that such an agent is not a moral agent, and the designer bears full responsibility for its behavior. We also demonstrate that even if there is an artificial agent with a fully modifiable table capable of learning* and intentionality* that meets the conditions set by Floridi and Sanders for ascribing moral agency to an artificial agent, the designer retains strong moral responsibility. (shrink)
What is nontrivial digital computation? It is the processing of discrete data through discrete state transitions in accordance with finite instructional information. The motivation for our account is that many previous attempts to answer this question are inadequate, and also that this account accords with the common intuition that digital computation is a type of information processing. We use the notion of reachability in a graph to defend this characterization in memory-based systems and underscore the importance of instructional information for (...) digital computation. We argue that our account evaluates positively against adequacy criteria for accounts of computation. (shrink)
As software developers design artificial agents , they often have to wrestle with complex issues, issues that have philosophical and ethical importance. This paper addresses two key questions at the intersection of philosophy and technology: What is deception? And when is it permissible for the developer of a computer artifact to be deceptive in the artifact’s development? While exploring these questions from the perspective of a software developer, we examine the relationship of deception and trust. Are developers using deception to (...) gain our trust? Is trust generated through technological “enchantment” warranted? Next, we investigate more complex questions of how deception that involves AAs differs from deception that only involves humans. Finally, we analyze the role and responsibility of developers in trust situations that involve both humans and AAs. (shrink)
This paper analyzes certain technical details of Floridi’s Theory of Strongly Semantic Information. It provides a clarification regarding desirable properties of degrees of informativeness functions by rejecting three of Floridi’s original constraints and proposing a replacement constraint. Finally, the paper briefly explores the notion of quantities of inaccuracy and shows an analysis that mimics Floridi’s analysis of quantities of vacuity.
In their important paper “Autonomous Agents”, Floridi and Sanders use “levels of abstraction” to argue that computers are or may soon be moral agents. In this paper we use the same levels of abstraction to illuminate differences between human moral agents and computers. In their paper, Floridi and Sanders contributed definitions of autonomy, moral accountability and responsibility, but they have not explored deeply some essential questions that need to be answered by computer scientists who design artificial agents. One such question (...) is, “Can an artificial agent that changes its own programming become so autonomous that the original designer is no longer responsible for the behavior of the artificial agent?” To explore this question, we distinguish between LoA1 (the user view) and LoA2 (the designer view) by exploring the concepts of unmodifiable, modifiable and fully modifiable tables that control artificial agents. We demonstrate that an agent with an unmodifiable table, when viewed at LoA2, distinguishes an artificial agent from a human one. This distinction supports our first counter-claim to Floridi and Sanders, namely, that such an agent is not a moral agent, and the designer bears full responsibility for its behavior. We also demonstrate that even if there is an artificial agent with a fully modifiable table capable of learning* and intentionality* that meets the conditions set by Floridi and Sanders for ascribing moral agency to an artificial agent, the designer retains strong moral responsibility. (shrink)
This commentary on Fresco's article "Information processing as an account of concrete digital computation" illuminates the two intertwined roles that the definition of the term "information" plays in Fresco's analysis. It provides analysis of the notion of actualizing control in information processing. The key point made is that not all control information in common computational devices cannot be processed.
This paper applies social-relational models of moral standing of robots to cases where the encounters between the robot and humans are relatively brief. Our analysis spans the spectrum of non-social robots to fully-social robots. We consider cases where the encounters are between a stranger and the robot and do not include its owner or operator. We conclude that the developers of robots that might be encountered by other people when the owner is not present cannot wash their hands of responsibility. (...) They must take care with how they develop the robot’s interface with people and take into account how that interface influences the social relationship between it and people, and, thus, the moral standing of the robot with each person it encounters. Furthermore, we claim that developers have responsibility for the impact social robots have on the quality of human social relationships. (shrink)
This paper analyzes certain technical details of Floridi’s Theory of Strongly Semantic Information. It provides a clarification regarding desirable properties of degrees of informativeness functions by rejecting three of Floridi’s original constraints and proposing a replacement constraint. Finally, the paper briefly explores the notion of quantities of inaccuracy and shows an analysis that mimics Floridi’s analysis of quantities of vacuity.
There is a growing literature on the concept of e-trust and on the feasibility and advisability of “trusting” artificial agents. In this paper we present an object-oriented model for thinking about trust in both face-to-face and digitally mediated environments. We review important recent contributions to this literature regarding e-trust in conjunction with presenting our model. We identify three important types of trust interactions and examine trust from the perspective of a software developer. Too often, the primary focus of research in (...) this area has been on the artificial agents and the humans they may encounter after they are deployed. We contend that the humans who design, implement, and deploy the artificial agents are crucial to any discussion of e-trust and to understanding the distinctions among the concepts of trust, e-trust and face-to-face trust. (shrink)
As exome and genome sequencing move into clinical application, questions surround how to elicit consent and handle potential return of individual genomic results. This study analyzes nine consent forms used in NIH-funded sequencing studies. Content analysis reveals considerable heterogeneity, including in defining results that may be returned, identifying potential benefits and risks of return, protecting privacy, addressing placement of results in the medical record, and data-sharing. In response to lack of consensus, we offer recommendations.
The federal Certificate of Confidentiality plays an important role in research on sensitive topics by authorizing researchers to refuse to disclose identifiable research data in response to subpoenas in any legal setting. However, there is little known about how effective Certificates are in practice. This article draws on our legal and empirical research on this topic to fill this information gap. It includes a description of the purpose of Certificates, their legislative and regulatory history, and a summary of the few (...) reported and unreported cases that have dealt with Certificates. In addition, we outline other statutory confidentiality protections, compare them to the Certificate's protections, and analyze some of the vulnerabilities of a Certificate's protections. This analysis allows us to make specific recommendations for strengthening the protections afforded to research data. (shrink)
To many who develop and use free software, the GNU General Public License represents an embodiment of the meaning of free software. In this paper we examine the definition and meaning of free software in the context of three events surrounding the GNU General Public License. We use a case involving the GPU software project to establish the importance of Freedom 0 in the meaning of free software. We analyze version 3 of the GNU General Public License and conclude that (...) although a credible case can be made that the added restrictions are consistent with the definition of free software, the case requires subtle arguments. Strong arguments against the added restrictions are less subtle, and may therefore be more convincing to many users and developers. We also analyze the Affero General Public License and conclude that it is inconsistent with the definition of free software. (shrink)
Incidental fndings of potential medical signifcance are seen in approximately 5-8 percent of asymptomatic subjects and 16 percent of symptomatic subjects participating in large computed tomography colonography studies, with the incidence varying further by CT acquisition technique. While most CTC research programs have a well-defned plan to detect and disclose IFs, such plans are largely communicated only verbally. Written consent documents should also inform subjects of how IFs of potential medical signifcance will be detected and reported in CTC research studies.
Background and objective: Assuming the hypothesis that the general practitioner can and should be a key player in making end-of-life decisions for hospitalised patients, perceptions of GPs’ role assigned to them by hospital doctors in making withdrawal decisions for such patients were surveyed.Design: Questionnaire survey.Setting: Urban and rural areas.Participants: GPs.Results: The response rate was 32.2% , and it was observed that 70.8% of respondents believed that their participation in withdrawal decisions for their hospitalised patients was essential, whereas 42.1% believed that (...) the hospital doctors were sufficiently skilled to make withdrawal decisions without input from the GPs. Most respondents were found to believe that they had the necessary skills and enough time to participate in withdrawal decisions. The last case of treatment withdrawal in hospital for one of their patients was described by 40% of respondents, of whom only 40.0% believed that they had participated actively in the decision process. The major factors in the multivariate analysis were the GP’s strong belief that his or her participation was essential , information on admission of the patient given to the GP by the hospital department , rural practice , visit to the patient dying in hospital and a request by the family to be kept informed about the patient .Conclusion: Strong interest was evinced among GPs regarding end-of-life issues, as well as considerable experience of patients dying at home. As GPs are more closely corrected to patients’ families, they may be a good choice for third-party intervention in making end-of-life decisions for hospitalised patients. (shrink)
This paper describes the Electronic Schoolbag, a digital workspace developed at the University of Savoie (France) and analyses its usages. This online environment is dedicated to the educational world: it offers pupils, students, teachers, school staff, or parents, personal and group workspaces in which individual or collaborative activities can take place. The flexibility of this software, allowing synchronous or asynchronous activities, lies in the “participation model”. This model allows groups themselves to describe and organise their activities. The architecture that permits (...) its implementation in the Electronic Schoolbag workspace is described. The study of the practices of the workspace is then presented. This requires different observation methods, according to the different procedures chosen: real practices provided by quantitative methods (analysis of the logs of the actions and questionnaires) and imagined practices provided by qualitative methods (semi-directive interviews). The results obtained from the university users allow us to assess the evolution of the usages for different periods and on different university sites. The observatory also lets us list the main uses of the Electronic Schoolbag for educative communication (collaborative vs. individual, informative vs. communicative). (shrink)
The use of the computer metaphor has led to the proposal of mind architecture (Pylyshyn 1984; Newell 1990) as a model of the organization of the mind. The dualist computational model, however, has, since the earliest days of psychological functionalism, required that the concepts mind architecture and brain architecture be remote from each other. The development of both connectionism and neurocomputational science, has sought to dispense with this dualism and provide general models of consciousness – a uniform cognitive architecture –, (...) which is in general reductionist, but which retains the computer metaphor. This paper examines, in the first place, the concepts of mind architecture and brain architecture, in order to evaluate the syntheses which have recently been offered. It then moves on to show how modifications which have been made to classical functionalist mind architectures, with the aim of making them compatible with brain architectures, are unable to resolve some of the most serious problems of functionalism. Some suggestions are given as to why it is not possible to relate mind structures and brain structures by using neurocomputational approaches, and finally the question is raised of the validity of reductionism in a theory which sets out to unite mind and brain architectures. (shrink)
