Trustworthy Agents Research Articles

Framing Effects on Judgments of Social Robots' (Im)Moral Behaviors.

Frames—discursive structures that make dimensions of a situation more or less salient—are understood to influence how people understand novel technologies. As technological agents are increasingly integrated into society, it becomes important to discover how native understandings (i.e., individual frames) of social robots are associated with how they are characterized by media, technology developers, and even the agents themselves (i.e., produced frames). Moreover, these individual and produced frames may influence the ways in which people see social robots as legitimate and trustworthy agents—especially in the face of (im)moral behavior. This three-study investigation begins to address this knowledge gap by 1) identifying individually held frames for explaining an android’s (im)moral behavior, and experimentally testing how produced frames prime judgments about an android’s morally ambiguous behavior in 2) mediated representations and 3) face-to-face exposures. Results indicate that people rely on discernible ground rules to explain social robot behaviors; these frames induced only limited effects on responsibility judgments of that robot’s morally ambiguous behavior. Evidence also suggests that technophobia-induced reactance may move people to reject a produced frame in favor of a divergent individual frame.

Trust-based decision-making framework for multiagent system

Multiagent system (MAS) is gaining popular attention in various applications especially with the rapid expansion of internet and computing. In general, MAS considers the collaboration of several agents working in the same environment together to achieve common objectives and also plays a critical role in supporting humans for attaining a comfortable life. Multiagent systems are used in numerous applications such as robotics, cloud computing, e-health, military, etc. Under such circumstances, trust among agents is crucial to ensure successful completion of a task especially when the task could only be completed through sharing of information and resources among the agents. This paper presents for the first time a study on the development of trust estimation models as a part of a decision-making framework that empirically computes and evaluates the trustworthiness of agents in MAS where the agent could choose to cooperate and collaborate only with the other trustworthy agents. The novelty of this study lies in the incorporation of beta reputation system into the Markov Games-based temporal difference learning framework and the development of a novel heuristic computation techniques adapted based on averaging methods. These models function as the integral part of the proposed MAS decision-making framework that chooses agents to collaborate with based on trustworthiness. The developed models and the framework are tested for their accuracy, efficiency and effectiveness using several analyses such as the time steps analysis, root-mean-square analysis and interaction analysis. The performances of the proposed models were better than the performances obtained using the trust models reported in the literature. Further real-world experiments are carried out to test the viability of the developed models in real-world applications.

Diffusion and persistence of false rumors in social media networks: implications of searchability on rumor self-correction on Twitter

Social media networks (SMN) such as Facebook and Twitter are infamous for facilitating the spread of potentially false rumors. Although it has been argued that SMN enable their users to identify and challenge false rumors through collective efforts to make sense of unverified information—a process typically referred to as self-correction—evidence suggests that users frequently fail to distinguish among rumors before they have been resolved. How users evaluate the veracity of a rumor can depend on the appraisals of others who participate in a conversation. Affordances such as the searchability of SMN, which enables users to learn about a rumor through dedicated search and query features rather than relying on interactions with their relational connections, might therefore affect the veracity judgments at which they arrive. This paper uses agent-based simulations to illustrate that searchability can hinder actors seeking to evaluate the trustworthiness of a rumor’s source and hence impede self-correction. The findings indicate that exchanges between related users can increase the likelihood that trustworthy agents transmit rumor messages, which can promote the propagation of useful information and corrective posts.

Radio Free Europe and Cold War Truth Games: Trustworthy Messages Beyond the Great Divide

This article examines the crucial relationship between truth, trust and information in the context of the Cold War. In particular, it analyses the work of Radio Free Europe (RFE) and its various networks with the Polish underground. First, using a concept of Michel Foucault, it’s focus lies on Cold War ‘truth games’ and the circulation of underground inscriptions beyond the Iron Curtain. Second, defining mobility and superimposed inscriptions as main axioms of power, it analyses how ‘immutable mobiles’, as defined by Bruno Latour, allowed RFE to become not only a broadcaster but also a ‘centre of calculation’. Starting from the micro level and RFE’s paperwork, it elaborates on the mediation of information beyond the ‘great divide’, and the heterogeneous ‘allies’ and networks that had to be mobilised to guarantee the stability of knowledge during the Cold War. Finally, in accordance with Latour’s Actor-Network Theory (ANT), it shall be shown how RFE and the quest for ‘trustworthy agents’ supported the Polish underground behind the Iron Curtain. The flat ontology of the ANT and its focus on the networking capacity of humans and technology are particularly suitable for the analysis of transnational and (infra-)structural developments.

Designing Trustworthy Product Recommendation Virtual Agents Operating Positive Emotion and Having Copious Amount of Knowledge.

Anthropomorphic agents used in online-shopping need to be trusted by users so that users feel comfortable buying products. In this paper, we propose a model for designing trustworthy agents by assuming two factors of trust, that is, emotion and knowledgeableness perceived. Our hypothesis is that when a user feels happy and perceives an agent as being highly knowledgeable, a high level of trust results between the user and agent. We conducted four experiments with participants to verify this hypothesis by preparing transition operators utilizing emotional contagion and knowledgeable utterances. As a result, we verified that users' internal states transitioned as expected and that the two factors significantly influenced their trust states.

Linking Trust to Trustworthiness

Trust is valuable when placed in trustworthy agents and activities, but damaging or costly when (mis)placed in untrustworthy agents and activities. So it is puzzling that much contemporary work on trust – such as that based on polling evidence – studies generic attitudes of trust in types of agent, institution or activity in complete abstraction from any account of trustworthiness. Information about others’ generic attitudes of trust or mistrust that take no account of evidence whether those attitudes are well or ill placed can offer little or no help for those who aim to place or refuse trust well. Information about attitudes is evidently useful to those who aim to influence those who hold them, which explains why polls about attitudes are popular with political parties, advertisers and other campaigning organisations. But where we aim not to influence others, but to place and refuse trust intelligently we must link trust to trustworthiness, and must focus on evidence of honesty, competence and reliability.

The Design of Trust Networks

One can use trust networks to find trustworthy information, people, products, and services on public networks. Hence, they have the potential to combine the advantages of search, recommendation systems, and social networks. But proper design and correct incentives are critical to the success of such networks. In this paper, I propose a trust network architecture that emphasizes simplicity and robustness. I propose a trust network with constrained trust relationships and design a decentralized search and recommendation process. I create both informational and monetary incentives to encourage joining the network, to investigate and discover other trustworthy agents, and to make commitments to them by trusting them, by insuring them, or even by directly investing in them. I show that making the correct judgments about trustworthiness of others and reporting it truthfully are the optimum strategies since they reward the agents both with information by providing access to more of the network and with monetary payments by paying them for their services as information intermediaries. The extensive income potential from the trust connections creates strong incentives to join the network, to create reliable trust connections, and to report them truthfully.

A Hybrid Artificial Reputation Model Involving Interaction Trust, Witness Information and the Trust Model to Calculate the Trust Value of Service Providers

Agent interaction in a community, such as the online buyer-seller scenario, is often uncertain, as when an agent comes in contact with other agents they initially know nothing about each other. Currently, many reputation models are developed that help service consumers select better service providers. Reputation models also help agents to make a decision on who they should trust and transact with in the future. These reputation models are either built on interaction trust that involves direct experience as a source of information or they are built upon witness information also known as word-of-mouth that involves the reports provided by others. Neither the interaction trust nor the witness information models alone succeed in such uncertain interactions. In this paper we propose a hybrid reputation model involving both interaction trust and witness information to address the shortcomings of existing reputation models when taken separately. A sample simulation is built to setup buyer-seller services and uncertain interactions. Experiments reveal that the hybrid approach leads to better selection of trustworthy agents where consumers select more reputable service providers, eventually helping consumers obtain more gains. Furthermore, the trust model developed is used in calculating trust values of service providers.

Trust in the CODA model: Opinion dynamics and the reliability of other agents

A model for the joint evolution of opinions and how much the agents trust each other is presented. The model is built using the framework of the Continuous Opinions and Discrete Actions (CODA) model. Instead of a fixed probability that the other agents will decide in the favor of the best choice, each agent considers that other agents might be one one of two types: trustworthy or useless. Trustworthy agents are considered more likely to be right than wrong, while the opposite holds for useless ones. Together with the opinion about the discussed issue, each agent also updates that probability for each one of the other agents it interacts withe probability each one it interacts with is of one type or the other. The dynamics of opinions and the evolution of the trust between the agents are studied. Clear evidences of the existence of two phases, one where strong polarization is observed and the other where a clear division is permanent and reinforced are observed. The transition seems signs of being a first-order transition, with a location dependent on both the parameters of the model and the initial conditions. This happens despite the fact that the trust network evolves much slower than the opinion on the central issue.

Party “capacity” in new democracies: how executive format affects the recruitment of presidents and prime ministers

Scholars and practitioners express concern that parties in “third wave” democracies are poorly developed, compared to parties in older democracies. We suggest that parties vary in their organizational “capacity”, focusing on parties' ability to select trustworthy executive agents. Capacity is higher where parties can vet potential executive talent by observing future leaders over time in the legislature – an increasingly available option as democracy matures. The key distinction in parties' use of this option lies in the delegation structure between a party and the executive. Parliamentary systems offer a clear line of delegation, which parties control. In presidential systems, parties must recruit executive candidates who can win a popular election, requiring characteristics that may not be well correlated with those that make them good party agents. As parliamentary democracy matures, we find a steady increase in prime ministers' average length of prior legislative service. For presidents, there is significantly weaker growth in prior legislative service. We also theorize about and investigate patterns in semi-presidential democracies. Our findings suggest that the institutional format of the executive is more important for party capacity in new democracies than the era in which a democracy was born.

Noblesse Oblige: how systems for accountability affect trust among professionals

Whereas professionals were once regarded as trustworthy agents whose actions stemmed from their ‘calling’ and social responsibility for the community, nowadays, numerous public scandals seem to have eroded the public’s trust. Institutions are continuously being challenged to account for their actions and trust in professionals seems to be replaced by various accountability systems – both calculative and narrative. Calculative accountability as the dominant means for building trust with outsiders seems to have reached its limits, and the question arises how accountability affects trust among professionals. Based on a qualitative case study using a structurational discourse analysis, we revealed how accountability affects several layers of trust among professionals. Calculative accountability served to showcase good practices for restoring trust outside the profession and retain autonomy. Narrative accountability, however, was perceived as essential for interprofessional trust and to genuinely discuss the real ethical matters. As distrust among professionals erupted about the issue of a quality control system, this threatened the trust necessary to benefit from narrative accountability. The proposed solution in the literature to implement both calculative and narrative accountability simultaneously, thus seems to run into a Catch-22 situation in practice.

Generalized framework for personalized recommendations in agent networks

An agent network can be modeled as a directed weighted graph whose vertices represent agents and edges represent a trust relationship between the agents. This article proposes a new recommendation approach, dubbed LocPat, which can recommend trustworthy agents to a requester in an agent network. We relate the recommendation problem to the graph similarity problem, and define the similarity measurement as a mutually reinforcing relation. We understand an agent as querying an agent network to which it belongs to generate personalized recommendations. We formulate a query into an agent network as a structure graph applied in a personalized manner that reflects the pattern of relationships centered on the requesting agent. We use this pattern as a basis for recommending an agent or object (a vertex in the graph). By calculating the vertex similarity between the agent network and a structure graph, we can produce a recommendation based on similarity scores that reflect both the link structure and the trust values on the edges. Our resulting approach is generic in that it can capture existing network-based approaches merely through the introduction of appropriate structure graphs. We evaluate different structure graphs with respect to two main kinds of settings, namely, social networks and ratings networks. Our experimental results show that our approach provides personalized and flexible recommendations effectively and efficiently based on local information.

A Trust Model for Supply Chain Management

Many real-world applications, such as Supply Chain Management (SCM), can be modeled using multi-agent systems. One shortcoming of current SCM models is that their trust models are ad hoc and do not have a strong theoretical basis. We propose a trust model for SCM that is grounded in probabilistic game theory. In this model, trust can be gained through direct interactions, and/or by asking for information from other trustworthy agents. We will use this model to simulate and study supply chain market behavior.

Trustworthiness and Economic Performance

Intrinsically trustworthy agents never cheat. A society's willingness to trust and the quality of its institutions have their origins in the intrinsic trustworthiness of its citizens. Trustworthiness is the basis for maximizing output in economic exchange and in explaining differences in standards of living around the world. We measure intrinsic trustworthiness with a question from the World Values Survey and estimate its effect using a sample of 60 countries. We find that trustworthiness is important for output per capita and that the effect of trust is likely to come from trustworthiness.

Trust, Social Capital, and Economic Development

Many argue that elements of a society’s norms, culture or social capital are central to understanding its development. However, these notions have been difficult to capture in economic models. Here we explore ap ossible role for ‘trustworthiness’ as corresponding to social capital. Individuals are trustworthy when they perform in accordance with promises, even if this does not maximize their payoffs. The usual focus on incentive structures in motivating behaviour plays no role here. Instead, we emphasize more deep-seated modes of behaviour and consider trustworthy agents being socialized to act as they do. To model this socialization, we borrow from a process of preference evolution pioneered by Bisin and Verdier (2001). The model developed endogenously accounts for social capital and explores its role in the process of economic development. It captures in a simple, formal way the interaction between social capital and the economy’s productive processes. The results obtained caution against rapid reform and provide an explanation for why late developing countries may not easily be able to transplant the modes of production that have proved useful in the West.

Trust, Social Capital and Economic Development

Many argue that elements of a society's norms, culture or social capital are central to understanding its development. However, these notions have been difficult to capture in economic models. Here, we argue that trustworthiness is the economically relevant component of a society's culture and, hence, comprises its social capital. Individuals are trustworthy when they perform actions they have promised, even if these do not maximize their payoffs. The usual focus on incentive structures in motivating behaviour plays no role here. Instead, we emphasize more deep-seated modes of behaviour and consider that trustworthy agents are socialized to act as they do. To model this socialization, we borrow from a relatively new process of preference evolution pioneered by Bisin and Verdier (2001). The model developed endogenously accounts for social capital and explores its role in the process of economic development. It captures in a simple, formal way the interaction between social capital and the economy's productive process. The results obtained caution against rapid reform, provide an explanation for why late developing countries cannot easily transplant the modes of production that have proved useful in the West, and suggest an explanation for the pattern of reform experiences in ex-communist countries.

Crowding out Trust: The Adverse Effect of Verification: An Experiment

Through an experiment, this study investigates the effects that verification has on honest traders. This paper demonstrates that by reducing the scope for trust, verification can have a negative effect on the behaviour of honest individuals. Specifically, the analysis shows that trustworthy agents are willing to punish or seek compensation from those that deprive them of trust opportunities through the use of verification. Regulatory implications are discussed.

The COMsumer manifesto: Empowering communities of consumers through the Internet

The coming economic era and precepts are emerging as totally different from industrial capitalism. We're learning that e-businesses and their networks destroy many of our basic concepts of production, marketing and distribution. Jeremy Rifkin [1] notes that in this "Age of Access" we are entering an era in which lifelong customer relationships are the ultimate commodities market. This provides a different, more positive, and perhaps more likely view. The Internet is changing business models and empowering consumers to create new communities that combine the power to aggregate rich sources of individually personalized data in real-time activities. Large-scale data aggregators are emerging to navigate and mediate info markets. While information records are proliferating, new standards for content capture and management are appearing. Most companies continue to hope they will control their customers' information assets. However, what if this is not true or becomes impossible? What if consumers decide to band together and control their own personal information? Are you ready to freely give your customers their data records? Are you prepared to live up to the COMsumer Manifesto? This article offers a disruptive antidote to the hierarchical, closed, supply-system, explicit, knowledge-driven, "We Know What You Want" data mine world where many customers feel powerless. This is a world well beyond 1999's "Net Worth" [2] and 2000's "The Cluetrain Manifesto" [3]. Infomediaries are not just trustworthy agents which sit between the vendor and the customer [2], and markets are not just conversations [3]. In this new world, communities sense needs, desires, and wishes for the future and create new data markets - to which organizations must respond or die! We are closing in on the "tipping point" [4] where COMsumers take complete control of their destiny by collectively owning their personal information assets.

The Rationality of Honesty in Debt Markets

The Rationality of Honesty in Debt Markets