Artificial intelligence tools used in employment decision-making cut across the multiple stages of job advertisements, shortlisting, interviews and hiring, and actual and potential bias can arise in each of these stages. One major challenge is to mitigate AI bias and promote fairness in opaque AI systems. This paper argues that the equal opportunity merit principle is an ethical approach for fair AI employment decision-making. Further, explainable AI can mitigate the opacity problem by placing greater emphasis on enhancing the understanding of (...) reasonable users and affected persons as to the AI output. Both the equal opportunity merit principle and explainable AI should be integrated in the design and implementation of AI employment decision-making systems so as to ensure, as far as possible, that the AI output is arrived at through a fair process. (shrink)

People’s comfort with and acceptability of artificial intelligence (AI) instantiations is a topic that has received little systematic study. This is surprising given the topic’s relevance to the design, deployment and even regulation of AI systems. To help fill in our knowledge base, we conducted mixed-methods analysis based on a survey of a representative sample of the US population (_N_ = 2254). Results show that there are two distinct social dimensions to comfort with AI: as a peer and as a (...) superior. For both dimensions, general and technological efficacy traits—locus of control, communication apprehension, robot phobia, and perceived technology competence—are strongly associated with acceptance of AI in various roles. Female and older respondents also were less comfortable with the idea of AI agents in various roles. A qualitative analysis of comments collected from respondents complemented our statistical approach. We conclude by exploring the implications of our research for AI acceptability in society. (shrink)

Mark Coeckelbergh (Int J Soc Robot 1:217–221, 2009) argues that robot ethics should investigate what interaction with robots can do to humans rather than focusing on the robot’s moral status. We should ask what robots do to our sociality and whether human–robot interaction can contribute to the human good and human flourishing. This paper extends Coeckelbergh’s call and investigate what it means to live with disembodied AI-powered agents. We address the following question: Can the human–AI interaction contribute to our moral (...) development? We present an empirically informed philosophical analysis of how the AI personal assistant Siri changes its users’ way of life, based on the responses obtained from 20 semi-directive individual interviews with Siri users. We identify changes in the users’ social interaction associated with the adoption of Siri. These changes include: (1) the indirect effect of reducing opportunities of human interaction, (2) the second-order effect of diminished expectations toward each other in a community, and (3) the acquired preference to obtain hassle-free interaction with Siri over human interaction. We examine them in relation to concerns that are voiced in the current debates over the rise of AI, namely the suspicion that humans could become overly reliant on AI (Danaher 2019) and the worry that social AI could impede on moral development (Fröding and Peterson, Ethics Inf Technol 23:207–214, 2012; Li, Ethics Inf Technol 23:543–550, 2021). We analyze the ethical costs that come from these changes in light of virtue ethics and address potential objections along the way. We end by offering directions for thinking about how to live with AI personal assistant while preserving favorable conditions for moral development. (shrink)

In this paper, we first classify different types of second opinions and evaluate the ethical and epistemological implications of providing those in a clinical context. Second, we discuss the issue of how artificial intelligent could replace the human cognitive labour of providing such second opinion and find that several AI reach the levels of accuracy and efficiency needed to clarify their use an urgent ethical issue. Third, we outline the normative conditions of how AI may be used as second opinion (...) in clinical processes, weighing the benefits of its efficiency against concerns of responsibility attribution. Fourth, we provide a ‘rule of disagreement’ that fulfils these conditions while retaining some of the benefits of expanding the use of AI-based decision support systems in clinical contexts. This is because the rule of disagreement proposes to use AI as much as possible, but retain the ability to use human second opinions to resolve disagreements between AI and physician-in-charge. Fifth, we discuss some counterarguments. (shrink)

C I Lewis showed up Down Under in 2005, in e-mails initiated by Allen Hazen of Melbourne. Their topic was the system Hazen called FL (a Funny Logic), axiomatized in passing in Lewis 1921. I show that FL is the system MEN of material equivalence with negation. But negation plays no special role in MEN. Symbolizing equivalence with → and defining ∼A inferentially as A→f, the theorems of MEN are just those of the underlying theory ME of pure material equivalence. (...) This accords with the treatment of negation in the Abelian l-group logic A of Meyer and Slaney (Abelian logic. Abstract, Journal of Symbolic Logic 46, 425–426, 1981), which also defines ∼A inferentially with no special conditions on f. The paper then concentrates on the pure implicational part AI of A, the simple logic of Abelian groups. The integers Z were known to be characteristic for AI, with every non-theorem B refutable mod some Zn for finite n. Noted here is that AI is pre-tabular, having the Scroggs property that every proper extension SI of AI, closed under substitution and detachment, has some finite Zn as its characteristic matrix. In particular FL is the extension for which n = 2 (Lewis, The structure of logic and its relation to other systems. The Journal of Philosophy 18, 505–516, 1921; Meyer and Slaney, Abelian logic. Abstract. Journal of Symbolic Logic 46, 425–426, 1981; This is an abstract of the much longer paper finally published in 1989 in G. G. Priest, R. Routley and J. Norman, eds., Paraconsistent logic: essays on the inconsistent, Philosophica Verlag, Munich, pp. 245–288, 1989). (shrink)

C I Lewis showed up Down Under in 2005, in e-mails initiated by Allen Hazen of Melbourne. Their topic was the system Hazen called FL (a Funny Logic), axiomatized in passing in Lewis 1921. I show that FL is the system MEN of material equivalence with negation. But negation plays no special role in MEN. Symbolizing equivalence with → and defining ∼A inferentially as A→f, the theorems of MEN are just those of the underlying theory ME of pure material equivalence. (...) This accords with the treatment of negation in the Abelian l-group logic A of Meyer and Slaney (Abelian logic. Abstract, Journal of Symbolic Logic 46, 425–426, 1981), which also defines ∼A inferentially with no special conditions on f. The paper then concentrates on the pure implicational part AI of A, the simple logic of Abelian groups. The integers Z were known to be characteristic for AI, with every non-theorem B refutable mod some Zn for finite n. Noted here is that AI is pre-tabular, having the Scroggs property that every proper extension SI of AI, closed under substitution and detachment, has some finite Zn as its characteristic matrix. In particular FL is the extension for which n = 2 (Lewis, The structure of logic and its relation to other systems. The Journal of Philosophy 18, 505–516, 1921; Meyer and Slaney, Abelian logic. Abstract. Journal of Symbolic Logic 46, 425–426, 1981; This is an abstract of the much longer paper finally published in 1989 in G. G. Priest, R. Routley and J. Norman, eds., Paraconsistent logic: essays on the inconsistent, Philosophica Verlag, Munich, pp. 245–288, 1989). (shrink)

The increased presence of medical AI in clinical use raises the ethical question which standard of explainability is required for an acceptable and responsible implementation of AI-based applications in medical contexts. In this paper, we elaborate on the emerging debate surrounding the standards of explainability for medical AI. For this, we first distinguish several goods explainability is usually considered to contribute to the use of AI in general, and medical AI in specific. Second, we propose to understand the value of (...) explainability relative to other available norms of explainable decision-making. Third, in pointing out that we usually accept heuristics and uses of bounded rationality for medical decision-making by physicians, we argue that the explainability of medical decisions should not be measured against an idealized diagnostic process, but according to practical considerations. We conclude, fourth, to resolve the issue of explainability-standards by relocating the issue to the AI’s certifiability and interpretability. (shrink)

Built-in decision thresholds for AI diagnostics are ethically problematic, as patients may differ in their attitudes about the risk of false-positive and false-negative results, which will require that clinicians assess patient values.

Background Machine learning (ML) is utilized increasingly in health care, and can pose harms to patients, clinicians, health systems, and the public. In response, regulators have proposed an approach that would shift more responsibility to ML developers for mitigating potential harms. To be effective, this approach requires ML developers to recognize, accept, and act on responsibility for mitigating harms. However, little is known regarding the perspectives of developers themselves regarding their obligations to mitigate harms.Methods We conducted 40 semi-structured interviews with (...) developers of ML predictive analytics applications for health care in the United States.Results Participants varied widely in their perspectives on personal responsibility and included examples of both moral engagement and disengagement, albeit in a variety of forms. While most (70%) of participants made a statement indicative of moral engagement, most of these statements reflected an awareness of moral issues, while only a subset of these included additional elements of engagement such as recognizing responsibility, alignment with personal values, addressing conflicts of interests, and opportunities for action. Further, we identified eight distinct categories of moral disengagement reflecting efforts to minimize potential harms or deflect personal responsibility for preventing or mitigating harms.Conclusions These findings suggest possible facilitators and barriers to the development of ethical ML that could act by encouraging moral engagement or discouraging moral disengagement. Regulatory approaches that depend on the ability of ML developers to recognize, accept, and act on responsibility for mitigating harms might have limited success without education and guidance for ML developers about the extent of their responsibilities and how to implement them. (shrink)

Artificial intelligence (AI) shares many generalizability challenges with psychology. But the fields publish differently. AI publishes fast, through rapid preprint sharing and conference publications. Psychology publishes more slowly, but creates integrative reviews and meta-analyses. We discuss the complementary advantages of each strategy, and suggest that incorporating both types of strategies could lead to more generalizable research in both fields.

McCradden et al. propose to close the “AI chasm” between algorithms and clinically meaningful application using the norms of evidence-based medicine and clinical research, with the rat...

Combining information from two or possibly several blocks of data is gaining increased attention and importance in several areas of science and industry. Typical examples can be found in chemistry, spectroscopy, metabolomics, genomics, systems biology and sensory science. Many methods and procedures have been proposed and used in practice. The area goes under different names: data integration, data fusion, multiblock analyses, multiset analyses and a few more. This book is an attempt to give an up-to-date treatment of the most used (...) and important methods within an important branch of the area; namely methods based on so-called components or latent variables. These methods have already obtained an enormous attention in for instance chemometrics, bioinformatics, machine learning and sensometrics and have proved to be important both for prediction and interpretation. The book is primarily a description of methodologies, but most of the methods will be illustrated by examples from the above-mentioned areas. The book is written such that both users of the methods as well as method developers will hopefully find sections of interest. In the end of the book there is a description of a software package developed particularly for the book. This package is freely available in R and covers many of the methods discussed. To distinguish the different type of methods from each other, the book is divided into five parts. Part I is introduction and preliminary concepts. Part II is the core of the book containing the main unsupervised and supervised methods. Part III deals with more complex structures and, finally, Part IV discusses alternative unsupervised and supervised methods. The book ends with Part V discussing the available software. Our recommendation for reading the book are as follows. A minimum read of the book would involve chapters 1, 2, 3, 5 and 7. Chapters 4, 6 and 8 are more specialized and chapters 9 and 10 contain methods we think are more advanced or less obvious to use. We feel privileged to have so many friendly colleagues who were willing to spend their time on helping us to improve the book by reading separate chapters. We would like to express our thanks to: Rasmus Bro, Margriet Hendriks, Ulf Indahl, Henk Kiers, Ingrid MaÌ⁽ge, Federico Marini, AÌ⁽smund Rinnan, Rosaria Romano, Lars Erik Solberg, Marieke Timmerman, Oliver Tomic, Johan Westerhuis and Barry Wise. Of course, the correctness of the final text is fully our responsibility! (shrink)

While the technologies that enable Artificial Intelligence (AI) continue to advance rapidly, there are increasing promises regarding AI’s beneficial outputs and concerns about the challenges of human–computer interaction in healthcare. To address these concerns, institutions have increasingly resorted to publishing AI guidelines for healthcare, aiming to align AI with ethical practices. However, guidelines as a form of written language can be analyzed to recognize the reciprocal links between its textual communication and underlying societal ideas. From this perspective, we conducted a (...) discourse analysis to understand how these guidelines construct, articulate, and frame ethics for AI in healthcare. We included eight guidelines and identified three prevalent and interwoven discourses: (1) AI is unavoidable and desirable; (2) AI needs to be guided with (some forms of) principles (3) trust in AI is instrumental and primary. These discourses signal an over-spillage of technical ideals to AI ethics, such as over-optimism and resulting hyper-criticism. This research provides insights into the underlying ideas present in AI guidelines and how guidelines influence the practice and alignment of AI with ethical, legal, and societal values expected to shape AI in healthcare. (shrink)

This paper explores the role and resolution of disagreements between physicians and their diagnostic AI-based decision support systems. With an ever-growing number of applications for these independently operating diagnostic tools, it becomes less and less clear what a physician ought to do in case their diagnosis is in faultless conflict with the results of the DSS. The consequences of such uncertainty can ultimately lead to effects detrimental to the intended purpose of such machines, e.g. by shifting the burden of proof (...) towards a physician. Thus, we require normative clarity for integrating these machines without affecting established, trusted, and relied upon workflows. In reconstructing different causes of conflicts between physicians and their AI-based tools—inspired by the approach of “meaningful human control” over autonomous systems and the challenges to resolve them—we will delineate normative conditions for “meaningful disagreements”. These incorporate the potential of DSS to take on more tasks and outline how the moral responsibility of a physician can be preserved in an increasingly automated clinical work environment. (shrink)

In a recent article, Hans Maes argues that examples drawn from contemporary visual art shed new light on the long-standing and seemingly intractable debate between Hypothetical Intentionalism (HI) and Moderate Actual Intentionalism (AI). He presents two test cases that, he argues, tilt the scale in favour of AI. In this paper I re-examine Maes's two test cases, and argue that neither succeeds as a test case. The first case fails because it confuses a relevant fact about the artwork with the (...) artist's intentions for the work. The second case fails because the work in question does not count as an utterance. The failure of Maes's examples suggests that the interpretive norms surrounding contemporary visual art cannot settle the debate between AI and HI. (shrink)

I tried to analyse the proceedings of heredity, methodologically separating the factors of “irritation and the reaction” upon it on the part of the living substance. The question was, to investigate the development of the hereditarily directed growth of the cells by “formative irritation” according to the conception of H.Driesch, that is to say in their homogeneousness with the biological reaction upon irritation on the part of the nutritive substances. It so became evident, that mechanism and vitalism have to co-operate (...) in separate sections of investigation: the first towards the substancially conceivable side, the latter towards an irrational order of ideas, as vitalism lays them down in “Entelechy“.Practically the same idea is expressed inSpemann's “Regenerations” by looking on the “Organisatoren” as substancial bearers of hereditary immaterial factors. The same thing results in all probability from an analysis ofPawlow's “conditioned Teflexes”.From all this results, that mechanism and vitalism ought to supplement one another in acting conjointly as to the conception of “totality” in the organic structure. In his “Holism”,Adolf Meyer, going beyond mechanism and vitalism, placed his conceptions of totality at the head and by “simplification” deducted the simple from the complicated.Hans Schmalfuss says: simplifying and complicating have each of them three sides, a substancial one, a planning-one and one constructed by thought. The planning side is the connecting link between the substancial side and the one constructed by thought and cannot be separated from either. He proves this by thorough argument in his “Substance and Life”. He is able to include naturallyvon Uexküll's “Planlehre”.All this corresponds to the adaptation of all that is living to the environment as belonging to a harmonious life of the individual. Even here the extreme mechanists suppose themselves able to draw upon the future and therefore remain implacable adversaries of vitalism. Perhaps there might be a reconciliation even here, if the order of ideas about an “All-life” as well as an “All-plan” combined with “Holism” should prove indispensable in scientifically enlightening the mystery of life.There follow a few words concerning the problem of heredity of qualities acquired, “erworbene Eigenschaften” .J'ai essayé d'analyser les procédés de l'hérédité en séparant méthodologiquement les facteurs de „l'irritation et ceux de la réaction” de la part de la substance vivante. Il s'agit ici du développement de la croissance des cellules par „l'irritation formative” d'après la conception de H.Driesch, cela veut dire d'après leur homogénéité avec la réaction biologique de la substance nutritive. Il se montra ainsi, que le mécanisme et le vitalisme doivent coopérer: le premier du côté substantiellement concevable, le dernier du côté de l'ordre des idées irrationelles telles que le vitalisme les expose dans „l'entéléchie”.Pratiquement, la même idée est exposée dans les „Régénérations” deSpemann, en regardant les „Organisatoren” comme porteurs des facteurs héréditaires et immatériels. Le même résultat se montre avec probabilité dans une analyse des „réflexes conditionels” dePawlow.De tout cela résulte, que le mécanisme et le vitalisme devraient suppléer l'un l'autre dans un travail commun quant à la notion de „totalité” dans la croissance organique. Dans son „Holisme”,Adolf Meyer, allant au delà du mécanisme et du vitalisme, a mis à la tête la conception de „totalité” et par la simplification il a déduit le simple du compliqué.Hans Schmalfuss dit: la simplification et la complication ont chacune trois côtés, un côté substantiel, un côté „planlich” et un côté provenant de la pensée. Le côté „planlich” est le lien entre le côté substantiel et celui provenant de la pensée et ne peut être séparé d'aucun des deux. Il prouve cela par un argument profond dans son livre „la Substance et la Vie”. Il inclut tout naturellement la „Planlehre” devon Uexküll.Tout cela correspond à l'adaptation de tout ce qui est vivant à son milieu, faisant ainsi part de la vie harmonieuse de l'individu. Il est vrai, que les extrêmes mécanistes croient pouvoir tirer sur l'avenir et c'est pour cette raison, qu'ils restent des adversaires implacables du vitalisme. Peut-être y aura-t-il une réconciliation si les ordres d'idées d'un „All-Leben” ainsi que d'un „All-Plan” combinés avec le „Holisme” se trouvent être indispensables à l'éclaircissement du mystère de la vie selon la science.Quelques mots suivent concernant le problème de l'hérédité des qualités acquises, „erworbene Eigenschaften”. (shrink)

Open peer commentary on the article “A Computational Constructivist Model as an Anticipatory Learning Mechanism for Coupled Agent–Environment Systems” by Filipo Studzinski Perotto. Upshot: Mainstream AI research largely addresses cognitive features as separate and unconnected. Instead of addressing cognitive growth in this same way – modeling it simply as one more such isolated feature and continuing to uphold a wrong-headed divide-and-conquer tradition – a constructivist approach should help unify many key phenomena such as anticipation, self-modeling, life-long learning, and recursive self-improvement. (...) Since this is likely to result in complex systems with unanticipated properties, all cognitive architecture researchers should aim to implement their ideas in full as running systems to be verified by experiment. Perotto’s paper falls short on both these points. (shrink)

The symbol grounding problem is concerned with the question of how the knowledge used in AI programs, expressed as tokens in one form or another or simply symbols, could be grounded to the outside world. By grounding the symbols, it is meant that the system will know the actual objects, events, or states of affairs in the world to which each symbol refers and thus be worldly-wise. Solving this problem, it was hoped, would enable the program to understand its own (...) action and hence be truly intelligent ). The problem becomes more acute after a challenge posed by Searle in his now famous Chinese Room Gedanken experiment. Searle argued that no AI programs can be said to understand or have other cognitive states, if all they do is formal symbol manipulation. (shrink)

This response offers considerable agreement with Anne Foerst's analysis in her essay “Cog, a Humanoid Robot, and the Question of the Image of God” (Zygon: Journal of Religion and Science 33 [March 1998]), yet endeavors to make her argument even more helpful. The response deals mainly with (1) the concept of symbol and the symbolic approach, (2) the symbolic description of a human being by artificial intelligence (AI) and by the theological symbol, “image of God” (imago dei), and (3) the (...) ensuing dialogue between scientists and theologians. (shrink)

Otto Neurath compared science to a ship at sea on which the sailors have to repair their vessel as they keep it afloat. Metascience is a ship of a similar sort. Do not worry. There are no repairs to report. But changes are being made at Metascience on an ongoing basis, even as we work to meet our production deadlines. With this, our second issue, we would like to announce some further changes with the journal.Ties Nijseen and Christi Lue who (...) have long been responsible for many of the matters related to Metascience are handing over the responsibilities to Lucy Fleet. Ties has been the Publishing Editor for Metascience since 2009, and Christi has supported him as editorial assistant since 2012. Ties and Christi are still with Springer, but they have taken on other projects. We would like to take this opportunity to thank them for their years of service to Metascience.Lucy will be our new Publishing Editor. She has worked for Springer for 11 years. She is, in addition, .. (shrink)

The paper analysis the strategies and concepts developed in the world in modern directions: innova- tive economy, digital economy, artificial intelligence, Industry 4.0 and others. The problem is to determine the initial fundamental parameters of order and their prospects in the global world, the definition and principles of artificial intel- ligence systems, its structure and important aspects and principles of future science and technology in analysis and synthesis based on synergetic approaches, innovative, information, converged technologies, taking into account the design (...) of future and viable and safe and sustainable development in the context of Industry 4.0, digital economy and Society 5.0. The general scheme of the integrated model of sustainable and socio-humanitarian development in the global system is proposed, which the authors call the noosphere model of sustainable and safe development as an important fundamental platform for the design and creation of AIS. In the XXI century education, science and innova- tive technologies becomes the «basis» of the spiritual and material reproduction of man, transforming the society of the XXI century into an educational society, ensures compliance with the law of advanced development of human quality, quality of social intelligence and quality of educational systems in society. noosphere evolution. Industry 4.0 is the next stage in the digitalization of production technologies with the accelerated introduction of new developments and cyberphysical systems, which is changing business models. Accordingly, Industry 4.0 as a separate segment of digital manufacturing technologies takes its place among other digital sectors. The analysis of the Strategies and Concepts developed in the world in modern directions: innovative economy, digital economy, artificial intelli- gence, Industry 4.0 and others is carried out in the work. The problem is to determine the initial fundamental para- meters of order and their prospects in the global world, the definition and principles of artificial intelligence systems, its structure and important aspects and principles of future science and technology in analysis and synthesis ba- sed on synergetic approaches, innovative, information, converged technologies, taking into account the design of future and viable and safe and sustainable development in the context of Industry 4.0, DE and Society 5.0. The main scientific and technological factors in the 21st century will be the development of: Artificial Intelligence, Nano Technologies, Bio-Media Technologies, Cognitive Technologies and Socio-Humanitarian Technologies. More specifically – this is the modern development of Innovative Technologies: «NBIC SH». Therefore, it is necessary to note the importance of taking into account the principle of I2K2+SH, ie: Intellectualization, Integration, Convergence, Coevolution and Socio- humanitarian technologies. It is important and necessary to take into account modern principles of designing a su- stainable and secure AIS and solving the problem of Harmonization between 2 worlds: Real and Virtual, especially in their hybridization. Our immediate future is a hybrid nonlinear world. Note that digitalization, artificial intelligence and robotics will be the main drivers of global economic growth by 2030E. An important paradigm and platform is the Integrated System: for Industry 4.0, DE and AI, taking into account the complex principle of I2K2+SH. If the transition of the Ukrainian economy to an innovative one does not take place in the next 3-5 years, Ukraine will remain in the backyard of civilization. (shrink)

Big data and artificial intelligence (“AI”) promise to transform virtually all aspects of biomedical research and health care (Matheny et al. 2019), through facilitation of drug development, diagno...