Our senses provide us with information about the world, but what exactly do they tell us? I argue that in order to optimally respond to sensory stimulations, an agent’s doxastic space may have an extra, “imaginary” dimension of possibility; perceptual experiences confer certainty on propositions in this dimension. To some extent, the resulting picture vindicates the old-fashioned empiricist idea that all empirical knowledge is based on a solid foundation of sense-datum propositions, but it avoids most of the problems traditionally associated (...) with that idea. The proposal might also explain why experiences appear to have a non-physical phenomenal character, even if the world is entirely physical. (shrink)

How should rational beliefs change over time? The standard Bayesian answer is: by conditionalization (a.k.a. Bayes’ Rule). But conditionalization is not an adequate rule for updating beliefs in “centred” propositions whose truth-value may itself change over time. In response, some have suggested that the objects of belief must be uncentred; others have suggested that beliefs in centred propositions are not subject to diachronic norms. Iargue that these views do not offer a satisfactory account of self-locating beliefs and their dynamics. A (...) third response is to replace conditionalization by a new norm that can deal with centred propositions. I critically survey anumber of new norms that have been proposed, and defend one particular approach. (shrink)

From 2010 to 2020, Indonesia’s tax-to-gross domestic product (GDP) ratio has been declining. A tax-to-GDP ratio trend of this magnitude indicates that the tax authority lacks the capacity to collect taxes. The tax administration system’s modernization utilizing information technology is thus deemed necessary. Artificial intelligence (AI) technology may serve as a solution to this issue. Using the theoretical frameworks of innovations in tax compliance, the cost of taxation, success factors for information technology governance (SFITG), and AI readiness, this study aims (...) to analyze the costs and benefits, the enablers and inhibitors, and the readiness of the government and related parties to apply AI to modernize the tax administration system in Indonesia. This study used qualitative approaches for the data’s collection and analysis. The data were obtained through a literature study and in-depth interviews. The findings show that AI application in the field of taxation can assist tax authorities in enforcing the law, provide taxpayers with convenience in fulfilling their tax obligations, improve justice for all taxpayers, and reduce tax compliance costs. The openness of Indonesia to technological developments, as evidenced by the AI National Strategy, is a supporting factor in the application of AI in Indonesia, particularly for the modernization of the tax administration system. The absence of specific regulations governing AI adoption, as well as a lack of human resources that can help the tax administration process, data, and infrastructure already support, are the impediments to implementing AI for the modernization of the tax administration system in Indonesia. (shrink)

I shall argue that software agents can be attributed cognitive states, since their behaviour can be best understood by adopting the intentional stance. These cognitive states are legally relevant when agents are delegated by their users to engage, without users’ review, in choices based on their the agents’ own knowledge. Consequently, both with regard to torts and to contracts, legal rules designed for humans can also be applied to software agents, even though the latter do not have rights and duties (...) of their own. The implications of this approach in different areas of the law are then discussed, in particular with regard to contracts, torts, and personality. (shrink)

As ever more advanced digital systems are created, it becomes increasingly likely that some of these systems will be digital minds, i.e. digital subjects of experience. With digital minds comes the risk of digital suffering. The problem of digital suffering is that of mitigating this risk. We argue that the problem of digital suffering is a high stakes moral problem and that formidable epistemic obstacles stand in the way of solving it. We then propose a strategy for solving it: Access (...) Monitor Prevent (AMP). AMP uses a ‘dancing qualia’ argument to link the functional states of certain digital systems to their experiences—this yields epistemic access to digital minds. With that access, we can prevent digital suffering by only creating advanced digital systems that we have such access to, monitoring their functional profiles, and preventing them from entering states with functional markers of suffering. After introducing and motivating AMP, we confront limitations it faces and identify some options for overcoming them. We argue that AMP fits especially well with—and so provides a moral reason to prioritize—one approach to creating such systems: whole brain emulation. We also contend that taking other paths to digital minds would be morally risky. (shrink)

This paper engages critically with anti-representationalist arguments pressed by prominent enactivists and their allies. The arguments in question are meant to show that the “as-such” and “job-description” problems constitute insurmountable challenges to causal-informational theories of mental content. In response to these challenges, a positive account of what makes a physical or computational structure a mental representation is proposed; the positive account is inspired partly by Dretske’s views about content and partly by the role of mental representations in contemporary cognitive scientific (...) modeling. (shrink)

Our aim in the present paper is to approach the nature of life from the perspective of autonomy, showing that this perspective can be helpful for overcoming the traditional Cartesian gap between the physical and cognitive domains. We first argue that, although the phenomenon of life manifests itself as highly complex and multidimensional, requiring various levels of description, individual organisms constitute the core of this multifarious phenomenology. Thereafter, our discussion focuses on the nature of the organization of individual living entities, (...) proposing autonomy as the main concept to grasp it. In the second part of the article we show how autonomy is also fundamental to explaining major evolutionary transitions, in an attempt to rethink evolution from the point of view of the organizational structure of the entities/organisms involved. This gives further support to the idea of autonomy not only as a key to understanding life in general but also the complex expressions of it that we observe on our planet. Finally, we suggest a possible general principle that underlies those evolutionary transitions, which allow for the open-ended redefinition of autonomous systems: namely, the relative dynamic decoupling that must be articulated among distinct parts, modules or modes of operation in these systems. (shrink)

Creating agents that proficiently interact with people is critical for many applications. Towards creating these agents, models are needed that effectively predict people's decisions in a variety of problems. To date, two approaches have been suggested to generally describe people's decision behavior. One approach creates a-priori predictions about people's behavior, either based on theoretical rational behavior or based on psychological models, including bounded rationality. A second type of approach focuses on creating models based exclusively on observations of people's behavior. At (...) the forefront of these types of methods are various machine learning algorithms. This paper explores how these two approaches can be compared and combined in different types of domains. In relatively simple domains, both psychological models and machine learning yield clear prediction models with nearly identical results. In more complex domains, the exact action predicted by psychological models is not even clear, and machine learning models are even less accurate. Nonetheless, we present a novel approach of creating hybrid methods that incorporate features from psychological models in conjunction with machine learning in order to create significantly improved models for predicting people's decisions. To demonstrate these claims, we present an overview of previous and new results, taken from representative domains ranging from a relatively simple optimization problem and complex domains such as negotiation and coordination without communication. (shrink)

There are many advantages to defining emotions as states elicited by reinforcers, with the states having a set of different functions. This approach leads towards an understanding of the nature of emotion, of its evolutionary adaptive value, and of many principles of brain design. It also leads towards a foundation for many of the processes that underlie evolutionary psychology and behavioral ecology. It is shown that recent as well as previous evidence implicates the amygdala and orbitofrontal cortex in positive as (...) well as negative emotions. The issue of why emotional states feel like something is part of the much larger problem of phenomenal consciousness. It is argued that thinking about one's own thoughts would have adaptive value by enabling first order linguistic thoughts to be corrected. It is suggested that reflecting on and correcting one's own thoughts and plans would feel like something, and that phenomenal consciousness may occur when this type of monitoring process is taking place. (shrink)

Artificial intelligence has dramatically changed the way organizations communicate, understand, and interact with their potential consumers. In the context of this trend, the ethical considerations of advertising when applying AI should be the core question for marketers. This paper discusses six dominant algorithmic purchase decision pathways that align with ethical philosophies for online customers when buying a product/goods. The six ethical positions include: ethical egoism, deontology, relativist, utilitarianism, virtue ethics, and ethics of care. Furthermore, this paper launches an “intelligent advertising” (...) AI theme by examining its present and future as well as identifying the key phases of intelligent advertising. Several research questions are offered to guide future research on intelligent advertising to benefit ethical AI decision-making. Finally, several areas that can be widely applied to ethical intelligent advertising are suggested for future research. (shrink)

Fodor advocates a view of cognitive processes as computations defined over the language of thought (or Mentalese). Even among those who endorse Mentalese, considerable controversy surrounds its representational format. What semantically relevant structure should scientific psychology attribute to Mentalese symbols? Researchers commonly emphasize logical structure, akin to that displayed by predicate calculus sentences. To counteract this tendency, I discuss computational models of navigation drawn from probabilistic robotics. These models involve computations defined over cognitive maps, which have geometric rather than logical (...) structure. They thereby demonstrate the possibility of rational cognitive processes in an exclusively non-logical representational medium. Furthermore, they offer much promise for the empirical study of animal navigation. (shrink)

In this paper, we explore and describe what is needed to allow connected and automated vehicles to break traffic rules in order to minimise road safety risk and to operate with appropriate transparency. Reviewing current traffic rules with particular reference to two driving situations, we illustrate why current traffic rules are not suitable for CAVs and why making new traffic rules specifically for CAVs would be inappropriate. In defining an alternative approach to achieving safe CAV driving behaviours, we describe the (...) use of ethical goal functions as part of hybrid AI systems, suggesting that functions should be defined by governmental bodies with input from citizens and stakeholders. Ethical goal functions for CAVs would enable developers to optimise driving behaviours for safety under conditions of uncertainty whilst allowing for differentiation of products according to brand values. Such functions can differ between regions according to preferences for safety behaviours within that region and can be updated over time, responding to continual socio-technological feedback loops. We conclude that defining ethical goal functions is an urgent and necessary step from governmental bodies to enable the safe and transparent operation of CAVs and accelerate the reduction in road casualties they promise to achieve. (shrink)

The understanding of decision-making systems has come together in recent years to form a unified theory of decision-making in the mammalian brain as arising from multiple, interacting systems (a planning system, a habit system, and a situation-recognition system). This unified decision-making system has multiple potential access points through which it can be driven to make maladaptive choices, particularly choices that entail seeking of certain drugs or behaviors. We identify 10 key vulnerabilities in the system: (1) moving away from homeostasis, (2) (...) changing allostatic set points, (3) euphorigenic signals, (4) overvaluation in the planning system, (5) incorrect search of situation-action-outcome relationships, (6) misclassification of situations, (7) overvaluation in the habit system, (8) a mismatch in the balance of the two decision systems, (9) over-fast discounting processes, and (10) changed learning rates. These vulnerabilities provide a taxonomy of potential problems with decision-making systems. Although each vulnerability can drive an agent to return to the addictive choice, each vulnerability also implies a characteristic symptomology. Different drugs, different behaviors, and different individuals are likely to access different vulnerabilities. This has implications for an individual's susceptibility to addiction and the transition to addiction, for the potential for relapse, and for the potential for treatment. (shrink)

Development of decision-support and intelligent agent systems necessitates mathematical descriptions of uncertainty and fuzziness in order to model vagueness. This paper seeks to present an outline of Peirce’s triadic logic as a practical new way to model vagueness in the context of artificial intelligence. Charles Sanders Peirce was an American scientist–philosopher and a great logician whose triadic logic is a culmination of the study of semiotics and the mathematical study of anti-Cantorean model of continuity and infinitesimals. After presenting Peircean semiotics (...) within AI perspective, a mathematical formulation of a Peircean triadic set is given in relationship with classical and fuzzy sets. Using basic logical operators, all possible respective implication operators, bi-equivalence operators, valid rules of inference, and associative, distributive and commutative logical properties are derived and verified through the truth function approach. In order to suggest practical directions, aggregation operators for Peirce’s triadic logic have been formulated. A mathematical formulation of a medical diagnostic problem and ER diagram of a library management system using Peirce’s triadic relation show potential for further applications of the proposed triadic set and triadic logic. Alongside, a classical AI game—The Wumpus World—is implemented to show practical efficacy in comparison with binary implementation. Besides giving some preliminary formulations for trichotomous set theory and definition of finite automaton, development of hybrid architectures for intelligent agents and evolutionary computations are discussed as potential practical avenues for Peirce’s triadic logic. (shrink)

Predictive processing framework has found wide applications in cognitive science and philosophy. It is an attractive candidate for a unified account of the mind in which perception, action, and cognition fit together in a single model. However, PP cannot claim this role if it fails to accommodate an essential part of cognition—conceptual thought. Recently, Williams argued that PP struggles to address at least two of thought’s core properties—generality and rich compositionality. In this paper, I show that neither necessarily presents a (...) problem for PP. In particular, I argue that because we do not have access to cognitive processes but only to their conscious manifestations, compositionality may be a manifest property of thought, rather than a feature of the thinking process, and result from the interplay of thinking and language. Pace Williams, both of these capacities, constituting parts of a complex and multifarious cognitive system, may be fully based on the architectural principles of PP. Under the assumption that language presents a subsystem separate from conceptual thought, I sketch out one possible way for PP to accommodate both generality and rich compositionality. (shrink)

Mental representations remain the central posits of psychology after many decades of scrutiny. However, there is no consensus about the representational format(s) of biological cognition. This paper provides a survey of evidence from computational cognitive psychology, perceptual psychology, developmental psychology, comparative psychology, and social psychology, and concludes that one type of format that routinely crops up is the language-of-thought (LoT). We outline six core properties of LoTs: (i) discrete constituents; (ii) role-filler independence; (iii) predicate–argument structure; (iv) logical operators; (v) inferential (...) promiscuity; and (vi) abstract content. These properties cluster together throughout cognitive science. Bayesian computational modeling, compositional features of object perception, complex infant and animal reasoning, and automatic, intuitive cognition in adults all implicate LoT-like structures. Instead of regarding LoT as a relic of the previous century, researchers in cognitive science and philosophy-of-mind must take seriously the explanatory breadth of LoT-based architectures. We grant that the mind may harbor many formats and architectures, including iconic and associative structures as well as deep-neural-network-like architectures. However, as computational/representational approaches to the mind continue to advance, classical compositional symbolic structures – that is, LoTs – only prove more flexible and well-supported over time. (shrink)

In this article, we discuss, from a complex systems perspective, possible implications of the rising dependency between autonomous human social/individual action, ubiquitous computing, and artificial intelligent systems. Investigation is made of ethical and political issues related to the application of ubiquitous computing resources to autonomous decision-making processes and to the enhancement of human cognition and action. We claim that without the feedback of fellow humans, which teaches us the consequences of our actions in real everyday life, the indiscriminate use of (...) ubiquitous computing in decision-making processes seems to be beyond the reach of any clear ethical control. We argue that the complex systems perspective may help us to foresee possible long-term consequences of our choices, in areas where human autonomous action can be directly affected by informational technologies. (shrink)

When should a scientific community be cognitively diverse? This article presents a model for studying how the heterogeneity of learning heuristics used by scientist agents affects the epistemic efficiency of a scientific community. By extending the epistemic landscapes modeling approach introduced by Weisberg and Muldoon, the article casts light on the micro-mechanisms mediating cognitive diversity, coordination, and problem-solving efficiency. The results suggest that social learning and cognitive diversity produce epistemic benefits only when the epistemic community is faced with problems of (...) sufficient difficulty. (shrink)

Although activity aimed at the construction of artificial intelligence started about 60 years ago however, contemporary intelligent systems are effective in very narrow domains only. One of the reasons for this situation appears to be serious problems in the theory of intelligence. Intelligence is a characteristic of goal-directed systems and two classes of goal-directed systems can be derived from observations on animals and humans, one class is systems with innately and jointly determined goals and means. The other class contains systems (...) that are able to construct arbitrary goals and means. It is suggested that the classes (that implicitly underlie most models of artificial intelligence) are insufficient to explain human goal-directed activity. A broader approach to goal-directed systems is considered. This approach suggests that humans are goal-directed systems that jointly synthesize arbitrary goals and means. Neural and psychological data favoring this hypothesis and its experimental validation are considered. A simple computer model based on the idea of joint synthesis to simulate goal-directed activity is presented. The usage of the idea of joint synthesis for the construction of artificial intelligence is discussed. (shrink)

Artificial intelligence receives attention in media as well as in academe and business. In media coverage and reporting, AI is predominantly described in contrasted terms, either as the ultimate solution to all human problems or the ultimate threat to all human existence. In academe, the focus of computer scientists is on developing systems that function, whereas philosophy scholars theorize about the implications of this functionality for human life. In the interface between technology and philosophy there is, however, one imperative aspect (...) of AI yet to be articulated: how do intelligent systems make inferences? We use the overarching concept “Artificial Intelligent Behaviour” which would include both cognition/processing and judgment/behaviour. We argue that due to the complexity and opacity of artificial inference, one needs to initiate systematic empirical studies of artificial intelligent behavior similar to what has previously been done to study human cognition, judgment and decision making. This will provide valid knowledge, outside of what current computer science methods can offer, about the judgments and decisions made by intelligent systems. Moreover, outside academe—in the public as well as the private sector—expertise in epistemology, critical thinking and reasoning are crucial to ensure human oversight of the artificial intelligent judgments and decisions that are made, because only competent human insight into AI-inference processes will ensure accountability. Such insights require systematic studies of AI-behaviour founded on the natural sciences and philosophy, as well as the employment of methodologies from the cognitive and behavioral sciences. (shrink)

One of the most impressive feats in robotics was the 2005 victory by a driverless Volkswagen Touareg in the DARPA Grand Challenge. This paper discusses what can be learned about the nature of representation from the car’s successful attempt to navigate the world. We review the hardware and software that it uses to interact with its environment, and describe how these techniques enable it to represent the world. We discuss robosemantics, the meaning of computational structures in robots. We argue that (...) the car constitutes a refutation of semantic arguments against the possibility of strong artificial intelligence. (shrink)

Why would we want to develop artificial human-like arithmetical intelligence, when computers already outperform humans in arithmetical calculations? Aside from arithmetic consisting of much more than mere calculations, one suggested reason is that AI research can help us explain the development of human arithmetical cognition. Here I argue that this question needs to be studied already in the context of basic, non-symbolic, numerical cognition. Analyzing recent machine learning research on artificial neural networks, I show how AI studies could potentially shed (...) light on the development of human numerical abilities, from the proto-arithmetical abilities of subitizing and estimating to counting procedures. Although the current results are far from conclusive and much more work is needed, I argue that AI research should be included in the interdisciplinary toolbox when we try to explain the development and character of numerical cognition and arithmetical intelligence. This makes it relevant also for the epistemology of mathematics. (shrink)

The objective of this work is to explore the notion of “action” and “agency” in artificial intelligence (AI). It employs a metaphysical notion of action and agency as an epistemological tool in the critique of the notion of “action” and “agency” in artificial intelligence. Hence, both a metaphysical and cognitive analysis is employed in the investigation of the quiddity and nature of action and agency per se, and how they are, by extension employed in the language and science of artificial (...) intelligence. The advent of the science of artificial intelligence and cognitive science, and the technological applications of artificial intelligence in the production of agents such as driverless cars and expert systems, have raised the question of moral, ethical and/or legal responsibility in AI agents. This has re-emphasized the importance of the philosophical discourse on the notions of action and agency, which in contemporary intellectual discourse are now perceived to be phenomena within the epistemic competence of the natural sciences. This paper argues that AI systems do not and cannot possess free agency and autonomy, thus cannot be morally and ethically responsible, hence, it recommends a socio-political response to the question of responsibility in AI. It is then the duty of individual nations, or the global community to define and enact policies on who shoulders the responsibility of actions executed by AIs. (shrink)

This article discusses three dimensions of creativity: metaphorical thinking; social interaction; and going beyond extrapolation in predictions. An overview of applications of neural networks in these three areas is offered. It is argued that the current reliance on the apparatus of statistical regression limits the scope of possibilities for neural networks in general, and in moving towards artificial creativity in particular. Artificial creativity may require revising some foundational principles on which neural networks are currently built.

Decision theorists disagree about how instrumentally rational agents, i.e., agents trying to achieve some goal, should behave in so-called Newcomb-like problems, with the main contenders being causal and evidential decision theory. Since the main goal of artificial intelligence research is to create machines that make instrumentally rational decisions, the disagreement pertains to this field. In addition to the more philosophical question of what the right decision theory is, the goal of AI poses the question of how to implement any given (...) decision theory in an AI. For example, how would one go about building an AI whose behavior matches evidential decision theory’s recommendations? Conversely, we can ask which decision theories describe the behavior of any existing AI design. In this paper, we study what decision theory an approval-directed agent, i.e., an agent whose goal it is to maximize the score it receives from an overseer, implements. If we assume that the overseer rewards the agent based on the expected value of some von Neumann–Morgenstern utility function, then such an approval-directed agent is guided by two decision theories: the one used by the agent to decide which action to choose in order to maximize the reward and the one used by the overseer to compute the expected utility of a chosen action. We show which of these two decision theories describes the agent’s behavior in which situations. (shrink)

We employ the support vector machine classifier, over different types of kernels, to investigate whether observable variables of individuals and their household information are able to describe their consumption decision of film at theaters in Brazil. Using a very big dataset of 340,000 individuals living in metropolitan areas of a whole large developing economy, we performed a Knowledge Discovery in Databases to classify the film consumers, which results in 80% instances correctly classified. To reduce the degrees of freedom for SVM (...) and to learn the more important determinants of film consumption, we apply the Linear Discriminant Analysis that allows us to identify the key determinants of this consumption. The main individual characteristics are age, education, income, and preferences for cultural goods. Regarding the main geographic characteristics, these are the timing of sample, population concentration, and supply of movie theaters. The results point to an ineffective policy for the sector at the time investigated. (shrink)

Is more information always better? Or are there some situations in which more information can make us worse off? Good (1967) argues that expected utility maximizers should always accept more information if the information is cost-free and relevant. But Good's argument presupposes that you are certain you will update by conditionalization. If we relax this assumption and allow agents to be uncertain about updating, these agents can be rationally required to reject free and relevant information. Since there are good reasons (...) to be uncertain about updating, rationality can require you to prefer ignorance. (shrink)

Ontology is the philosophical discipline which aims to understand how things in the world are divided into categories and how these categories are related together. This is exactly what information scientists aim for in creating structured, automated representations, called 'ontologies,' for managing information in fields such as science, government, industry, and healthcare. Currently, these systems are designed in a variety of different ways, so they cannot share data with one another. They are often idiosyncratically structured, accessible only to those who (...) created them, and unable to serve as inputs for automated reasoning. This volume shows, in a nontechnical way and using examples from medicine and biology, how the rigorous application of theories and insights from philosophical ontology can improve the ontologies upon which information management depends. (shrink)

This paper explores the important role of critical science, and in particular of post-colonial and decolonial theories, in understanding and shaping the ongoing advances in artificial intelligence. Artificial intelligence is viewed as amongst the technological advances that will reshape modern societies and their relations. While the design and deployment of systems that continually adapt holds the promise of far-reaching positive change, they simultaneously pose significant risks, especially to already vulnerable peoples. Values and power are central to this discussion. Decolonial theories (...) use historical hindsight to explain patterns of power that shape our intellectual, political, economic, and social world. By embedding a decolonial critical approach within its technical practice, AI communities can develop foresight and tactics that can better align research and technology development with established ethical principles, centring vulnerable peoples who continue to bear the brunt of negative impacts of innovation and scientific progress. We highlight problematic applications that are instances of coloniality, and using a decolonial lens, submit three tactics that can form a decolonial field of artificial intelligence: creating a critical technical practice of AI, seeking reverse tutelage and reverse pedagogies, and the renewal of affective and political communities. The years ahead will usher in a wave of new scientific breakthroughs and technologies driven by AI research, making it incumbent upon AI communities to strengthen the social contract through ethical foresight and the multiplicity of intellectual perspectives available to us, ultimately supporting future technologies that enable greater well-being, with the goal of beneficence and justice for all. (shrink)

This paper investigates the prospects of Rodney Brooks’ proposal for AI without representation. It turns out that the supposedly characteristic features of “new AI” (embodiment, situatedness, absence of reasoning, and absence of representation) are all present in conventional systems: “New AI” is just like old AI. Brooks proposal boils down to the architectural rejection of central control in intelligent agents—Which, however, turns out to be crucial. Some of more recent cognitive science suggests that we might do well to dispose of (...) the image of intelligent agents as central representation processors. If this paradigm shift is achieved, Brooks’ proposal for cognition without representation appears promising for full-blown intelligent agents—Though not for conscious agents. (shrink)

The proposed European Artificial Intelligence Act (AIA) is the first attempt to elaborate a general legal framework for AI carried out by any major global economy. As such, the AIA is likely to become a point of reference in the larger discourse on how AI systems can (and should) be regulated. In this article, we describe and discuss the two primary enforcement mechanisms proposed in the AIA: the _conformity assessments_ that providers of high-risk AI systems are expected to conduct, and (...) the _post-market monitoring plans_ that providers must establish to document the performance of high-risk AI systems throughout their lifetimes. We argue that the AIA can be interpreted as a proposal to establish a Europe-wide ecosystem for conducting AI auditing, albeit in other words. Our analysis offers two main contributions. First, by describing the enforcement mechanisms included in the AIA in terminology borrowed from existing literature on AI auditing, we help providers of AI systems understand how they can prove adherence to the requirements set out in the AIA in practice. Second, by examining the AIA from an auditing perspective, we seek to provide transferable lessons from previous research about how to refine further the regulatory approach outlined in the AIA. We conclude by highlighting seven aspects of the AIA where amendments (or simply clarifications) would be helpful. These include, above all, the need to translate vague concepts into verifiable criteria and to strengthen the institutional safeguards concerning conformity assessments based on internal checks. (shrink)

Within Non-monotonic Reasoning, numerous principles of causal reasoning have been proposed. Many of these principles have been viewed as desirable in formalisms that reason with causality, and have been widely adopted throughout the literature. We provide a critique of these principles, evaluate their suitability for characterising and formulating causal non-monotonic reasoning, and find that most are unsuitable. Further, we discuss a new approach to causal non-monotonic reasoning motivated by how humans typically reason with causality.

This paper compares current ways of modeling the inferential structure of practical reasoning arguments, and proposes a new approach in which it is regarded in a modular way. Practical reasoning is not simply seen as reasoning from a goal and a means to an action using the basic argumentation scheme. Instead, it is conceived as a complex structure of classificatory, evaluative, and practical inferences, which is formalized as a cluster of three types of distinct and interlocked argumentation schemes. Using two (...) real examples, we show how applying the three types of schemes to a cluster of practical argumentation allows an argument analyst to reconstruct the tacit premises presupposed and evaluate the argumentative reasoning steps involved. This approach will be shown to overcome the limitations of the existing models of practical reasoning arguments within the BDI and commitment theoretical frameworks, providing a useful tool for discourse analysis and other disciplines. In particular, applying this method brings to light the crucial role of classification in practical argumentation, showing how the ordering of values and preferences is only one of the possible areas of deep disagreement. (shrink)

Machine learning-based AI algorithms lack transparency. In this article, I offer an interpretation of AI’s explainability problem and highlight its ethical saliency. I try to make the case for the legal enforcement of a strong explainability requirement: human organizations which decide to automate decision-making should be legally obliged to demonstrate the capacity to explain and justify the algorithmic decisions that have an impact on the wellbeing, rights, and opportunities of those affected by the decisions. This legal duty can be derived (...) from the demands of Rawlsian public reason. In the second part of the paper, I try to show that the argument from the limitations of human cognition fails to get AI off the hook of public reason. Against a growing trend in AI ethics, my main argument is that the analogy between human minds and artificial neural networks fails because it suffers from an atomistic bias which makes it blind to the social and institutional dimension of human reasoning processes. I suggest that developing interpretive AI algorithms is not the only possible answer to the explainability problem; social and institutional answers are also available and in many cases more trustworthy than techno-scientific ones. (shrink)

As artificial intelligence (AI) continues to proliferate into every area of modern life, there is no doubt that society has to think deeply about the potential impact, whether negative or positive, that it will have. Whilst scholars recognise that AI can usher in a new era of personal, social and economic prosperity, they also warn of the potential for it to be misused towards the detriment of society. Deliberate strategies are therefore required to ensure that AI can be safely integrated (...) into society in a manner that would maximise the good for as many people as possible, whilst minimising the bad. One of the most urgent societal expectations of artificial agents is the need for them to behave in a manner that is morally relevant, i.e. to become artificial moral agents (AMAs). In this article, I will argue that exemplarism, an ethical theory based on virtue ethics, can be employed in the building of computationally rational AMAs with weak machine ethics. I further argue that three features of exemplarism, namely grounding in moral exemplars, meeting community expectations and practical simplicity, are crucial to its uniqueness and suitability for application in building AMAs that fit the ethos of AI4SG. (shrink)

Classic conceptions of model simplicity for machine learning are mainly based on the analysis of the structure of the model. Bayesian, Frequentist, information theoretic and expressive power concepts are the best known of them, which are reviewed in this work, along with their underlying assumptions and weaknesses. These approaches were developed before the advent of the Big Data deluge, which has overturned the importance of structural simplicity. The computational simplicity concept is presented, and it is argued that it is more (...) encompassing and closer to actual machine learning practices than the classic ones. In order to process the huge datasets which are commonplace nowadays, the computational complexity of the learning algorithm is the decisive factor to assess the viability of a machine learning strategy, while the classic accounts of simplicity play a surrogate role. Some of the desirable features of computational simplicity derive from its reliance on the learning system concept, which integrates key aspects of machine learning that are ignored by the classic concepts. Moreover, computational simplicity is directly associated with energy efficiency. In particular, the question of whether the maximum possibly achievable predictive accuracy should be attained, no matter the economic cost of the associated energy consumption pattern, is considered. (shrink)

The aim of this exploratory paper is to review an under-appreciated parallel between group agency and artificial intelligence. As both phenomena involve non-human goal-directed agents that can make a difference to the social world, they raise some similar moral and regulatory challenges, which require us to rethink some of our anthropocentric moral assumptions. Are humans always responsible for those entities’ actions, or could the entities bear responsibility themselves? Could the entities engage in normative reasoning? Could they even have rights and (...) a moral status? I will tentatively defend the (increasingly widely held) view that, under certain conditions, artificial intelligent systems, like corporate entities, might qualify as responsible moral agents and as holders of limited rights and legal personhood. I will further suggest that regulators should permit the use of autonomous artificial systems in high-stakes settings only if they are engineered to function as moral (not just intentional) agents and/or there is some liability-transfer arrangement in place. I will finally raise the possibility that if artificial systems ever became phenomenally conscious, there might be a case for extending a stronger moral status to them, but argue that, as of now, this remains very hypothetical. (shrink)

In this paper, I propose a logical-cognitive approach to argumentation and advocate an idea that argumentation presupposes that intelligent agents engaged in it are cognitively diverse. My approach to argumentation allows drawing distinctions between justification, conviction and persuasion as its different kinds. In justification agents seek to verify weak or strong coherency of an agent’s position in a dialogue. In conviction they argue to modify their partner’s position by means of demonstrating weak or strong cogency of their positions before a (...) ‘rational judge’. My approach to argumentation employs a ‘light’ version of Dung’s abstract argumentative frameworks. It is based on Stich’s idea of agents’ cognitive diversity the epistemic aspect of which is argued to be close to Pavilionis’s conception of meaning continuum. To illustrate my contributions I use an example based on the Kitchen Debate (1959) between Khrushchev and Nixon. (shrink)

The increasing use of Artificial Intelligence for making decisions in public affairs has sparked a lively debate on the benefits and potential harms of self-learning technologies, ranging from the hopes of fully informed and objectively taken decisions to fear for the destruction of mankind. To prevent the negative outcomes and to achieve accountable systems, many have argued that we need to open up the “black box” of AI decision-making and make it more transparent. Whereas this debate has primarily focused on (...) how transparency can secure high-quality, fair, and reliable decisions, far less attention has been devoted to the role of transparency when it comes to how the general public come to perceive AI decision-making as legitimate and worthy of acceptance. Since relying on coercion is not only normatively problematic but also costly and highly inefficient, perceived legitimacy is fundamental to the democratic system. This paper discusses how transparency in and about AI decision-making can affect the public’s perception of the legitimacy of decisions and decision-makers and produce a framework for analyzing these questions. We argue that a limited form of transparency that focuses on providing justifications for decisions has the potential to provide sufficient ground for perceived legitimacy without producing the harms full transparency would bring. (shrink)

There is enormous interest in using artificial intelligence (AI) in health care contexts. But before AI can be used in such settings, we need to make sure that AI researchers and organizations follow appropriate ethical frameworks and guidelines when developing these technologies. In recent years, a great number of ethical frameworks for AI have been proposed. However, these frameworks have tended to be abstract and not explain what grounds and justifies their recommendations and how one should use these recommendations in (...) practice. In this paper, I propose an AI ethics framework that is grounded in substantive, human rights theory and one that can help us address these questions. (shrink)

Because intelligent agents employ physically embodied cognitive systems to reason about the world, their cognitive abilities are constrained by the laws of physics. Scientists have used digital computers to develop and validate theories of physically embodied cognition. Computational theories of intelligence have advanced our understanding of the nature of intelligence and have yielded practically useful systems exhibiting some degree of intelligence. However, the view of cognition as algorithms running on digital computers rests on implicit assumptions about the physical world that (...) are incorrect. Recently, the view is emerging of computing systems as goal-directed agents, evolving during problem solving toward improved world models and better task performance. A full realization of this vision requires a new logic for computing that incorporates learning from experience as an intrinsic part of the logic, and that permits full exploitation of the quantum nature of the physical world. This paper proposes a theory of physically embodied cognitive agents founded upon first-order logic, Bayesian decision theory, and quantum physics. An abstract architecture for a physically embodied cognitive agent is presented. The cognitive aspect is represented as a Bayesian decision theoretic agent; the physical aspect is represented as a quantum process; and these aspects are related through von Neumann’s principle of psycho-physical parallelism. Alternative metaphysical positions regarding the meaning of quantum probabilities and the role of efficacious choices by agents are discussed in relation to the abstract agent architecture. The concepts are illustrated with an extended example from the domain of science fiction. (shrink)

The basic Bayesian model of credence states, where each individual’s belief state is represented by a single probability measure, has been criticized as psychologically implausible, unable to represent the intuitive distinction between precise and imprecise probabilities, and normatively unjustifiable due to a need to adopt arbitrary, unmotivated priors. These arguments are often used to motivate a model on which imprecise credal states are represented by sets of probability measures. I connect this debate with recent work in Bayesian cognitive science, where (...) probabilistic models are typically provided with explicit hierarchical structure. Hierarchical Bayesian models are immune to many classic arguments against single-measure models. They represent grades of imprecision in probability assignments automatically, have strong psychological motivation, and can be normatively justified even when certain arbitrary decisions are required. In addition, hierarchical models show much more plausible learning behavior than flat representations in terms of sets of measures, which—on standard assumptions about update—rule out simple cases of learning from a starting point of total ignorance. (shrink)

Artificial intelligence (AI) research enjoyed an initial period of enthusiasm in the 1970s and 80s. But this enthusiasm was tempered by a long interlude of frustration when genuinely useful AI applications failed to be forthcoming. Today, we are experiencing once again a period of enthusiasm, fired above all by the successes of the technology of deep neural networks or deep machine learning. In this paper we draw attention to what we take to be serious problems underlying current views of artificial (...) intelligence encouraged by these successes, especially in the domain of language processing. We then show an alternative approach to language-centric AI, in which we identify a role for philosophy. (shrink)

Recent progress in artificial intelligence has renewed interest in building systems that learn and think like people. Many advances have come from using deep neural networks trained end-to-end in tasks such as object recognition, video games, and board games, achieving performance that equals or even beats that of humans in some respects. Despite their biological inspiration and performance achievements, these systems differ from human intelligence in crucial ways. We review progress in cognitive science suggesting that truly human-like learning and thinking (...) machines will have to reach beyond current engineering trends in both what they learn and how they learn it. Specifically, we argue that these machines should build causal models of the world that support explanation and understanding, rather than merely solving pattern recognition problems; ground learning in intuitive theories of physics and psychology to support and enrich the knowledge that is learned; and harness compositionality and learning-to-learn to rapidly acquire and generalize knowledge to new tasks and situations. We suggest concrete challenges and promising routes toward these goals that can combine the strengths of recent neural network advances with more structured cognitive models. (shrink)

Criminal liability for acts committed by AI systems has recently become a hot legal topic. This paper includes three different contributions. The first contribution is an analysis of the extent to which an AI system can satisfy the requirements for criminal liability: accomplishing an actus reus, having the corresponding mens rea, possessing the cognitive capacities needed for responsibility. The second contribution is a discussion of criminal activity accomplished by an AI entity, with reference to a recent case involving an online (...) bot, the Random Darknet Shopper. This discussion will provide the context for the analysis of commonalities and differences between criminal activities by humans and by artificial systems. The third contribution concerns the evaluation of different ways of addressing criminal activities by AI systems in a regulatory perspective. (shrink)

Artificial intelligence has been adopted in a wide range of domains. This shows the imperative need to contribute to making citizens insightful actors in debates and decisions involving the adoption of AI mechanisms. Currently, existing approaches to the teaching of basic AI concepts through programming treat machine intelligence as an external element/module. After being trained, that external module is coupled to the main application. Combining block-based programming and WiSARD weightless artificial neural networks, this article presents the conceptualization and design of (...) a new methodology, AI from concrete to abstract, to endow general people with a minimum understanding of what AI means. The main strategy adopted was to include AI training and classification primitives as blocks that compose an otherwise conventional computer program. This way, the programmer can use these blocks as he/she uses other programming constructs. In order to achieve this purpose, we also propose BlockWiSARD, a block-based programming environment designed to promote the demystification of artificial intelligence via practical activities related to the development of learning machines, as well as through the observation of their learning process. As a beneficial side effect of BlockWiSARD, the difference between a program capable of learning from data and a conventional computer program becomes more evident. In addition, the simplicity of the WiSARD weightless artificial neural network model enables easy visualization and understanding of training and classification tasks internal realization. (shrink)