Socially useful artificial intelligence

Nurses are increasingly having to deal with complex situations involving sensitive face-to-face interpersonal skills at times of distress and complex interpersonal skills with a range of different people in the context of organizational change. These two types of situation represent the ends of a continuum of interpersonal skills in which the social problem-solving process can be useful. The continuum is one of social complexity. Relatively simple social situations are those in which few people are involved and their reasons for being involved are quite clear. Complex situations are those in which very many people are involved and the nature of their involvement varies and may or may not be clear. The complexity of the situation is not defined in terms of the complexity of the issue under consideration. As we shall see, simple situations can involve highly complicated issues.

Even though there exists limited knowledge on how exactly students acquire problem-solving competences, researchers agree that adaptive teaching interventions have the potential to support students‘ autonomous problem-solving processes. However, most recent research aims at analyzing the characteristics of teaching interventions rather than the interventions’ effects on the students’ problem-solving process. The study in this paper addresses this research gap by focusing not only on the teaching interventions themselves, but also on the students’ collaborative problem-solving processes just before and just after the interventions. The aim of the study is to analyze the interventions‘ effect on the learners’ integrated problem-solving processes.

Utilising D'Zurilla's (1986, 1990) transactional social problem solving model as the theoretical framework, the present study sought to examine the dynamics of the social problem solving process in relation to intraindividual experiences of events, mood, and physical health in daily life. The study incorporated both idiographic and nomothetic strategies, combining a daily diary approach within a prospective design. As such, each individual's experience of events, mood, and health were assessed on a daily basis, and the prospective relations between social problem solving and these day-to-day experiences examined. Participants' (university students) completed a measure of social problem solving at baseline followed, approximately 5 weeks later, by daily self-reports of mood (positive & negative), events (hassles & uplifts), and physical health (health status & URI symptoms) for a period of 14 days. The data were analysed using multilevel hierarchical modelling. Days on which individuals had greater than their average hassles were associated with higher negative and lower positive mood (i.e. domain specific & cross-over effects), whilst greater uplifts were linked to positive mood only. With regard to health, individuals reported worse status and more URI symptoms on days with greater than their average negative (hassles, negative mood) but not positive influences (uplifts, positive mood). Both social problem solving orientations and skills dimensions were associated prospectively with daily health outcomes, and these relationships were not moderated by day-to-day experiences of mood or events. The data clarify the importance of social problem solving to within-person daily event, mood, and health experiences. The implications of these findings for contemporary transactional social problem solving models are discussed.

This article argues that a relatively novel methodology called multi-agent artificial intelligence modeling can play an important role in helping scholars fulfill eight desiderata for a “good” social scientific theory (conceptual clarity, logical consistency, empirical groundedness, parsimony, generativity, testability, insightfulness, and usefulness). The unique contributions of this methodology include its use of psychologically realistic agents in sociologically realistic networks that interact with each other and their simulated environment within an “artificial society.” These simulation tools utilize artificial intelligence in a way that enables scholars to causally generate the emergence of macro -level societal phenomena of interest from the micro -level behaviors and meso -level interactions of simulated agents that represent real world populations. These social digital twins can also integrate multiple social theories within a single causal architecture, providing a unique opportunity for the revolutionizing of our best theoretical frameworks.

This opening chapter introduces a systems approach to exploring the intersection of artificial intelligence (AI) and education, focusing on how AI can be harnessed to meaningfully transform basic education systems, especially in the Global South. The authors define education systems as the complex ecology of institutions—schools, curricula, teachers, and governance mechanisms—that collectively shape learning opportunities for the majority of the world’s children. AI is conceptualized as technologies that can simulate human intelligence to undertake tasks such as communication, reasoning, and learning. Rather than assuming inherent benefits, the chapter adopts an open-minded yet cautious stance, emphasizing past disappointments with educational technology and the importance of distinguishing between marginal improvements and true systemic transformation. The authors highlight three central domains for examining AI’s contributions in education: fostering AI literacy, enhancing the effectiveness of educational delivery, and improving the relevance of school curricula to evolving societal and workforce demands. The chapter places a particular emphasis on the Global South, where resource constraints, demographic realities, and system fragility pose distinctive challenges—and where the bulk of the world’s youth reside. The argument is made that AI’s role cannot be understood in isolation; instead, a systems perspective is necessary, one that acknowledges the dynamic interplay between curriculum, teachers, assessments, and broader organizational, policy, and social contexts. A systems approach requires adapting AI solutions to local priorities and realities, ensuring that technology does not exacerbate existing inequalities or distract from foundational challenges—such as literacy and numeracy—while also seizing opportunities to leapfrog traditional constraints. By integrating complexity science concepts such as feedback loops and emergence, the authors illustrate how interventions in one part of the system can have far-reaching, sometimes unpredictable, consequences. The chapter further outlines the rationale for the book’s focus, exploring why educational institutions must contend with AI in light of labor market transformations, emerging needs for human and ethical skills and twenty-first century skills, and the potential for AI-enabled organizational improvements. It reviews current definitions and applications of AI in education, addressing both its potential and its limitations. The authors preview subsequent analysis on integrating AI into curricula to build AI literacy, leveraging AI to improve student learning and school functioning, and mitigating the risks of displacement and inequality. Throughout, the text calls for multi-level, evidence-based, ethically conscious policies as well as collaboration across government, industry, and educational stakeholders. Ultimately, the chapter sets the stage for a nuanced investigation into whether—and how—AI can deliver on its promise to transform education systems in meaningful, equitable, and sustainable ways.

Objectives: This study aimed to achieve a main goal: analyzing the aspects of focusing on models from social theories to interpret the social dimensions of artificial intelligence and the challenges associated with the evolution of its technologies. Methods: This study adopted the inductive approach with the aim of exploring and analyzing the social dimensions of artificial intelligence within the context of relevant social theories. Results: The findings of the study concluded that the use of social theories provides a deep understanding of the social dimensions and impacts of artificial intelligence. Based on the theory of digital social space, the importance of addressing the impact of technologies to ensure more credible social interactions is highlighted. The theory of social action can be employed to explain changes in individual behavior and the redefinition of concepts of authority. According to social constructivism theory, artificial intelligence contributes to the formation of new digital identities, while globalization theory demonstrates its effect on global connectivity and cultural and economic challenges. On the other hand, social learning theory points to its role in shaping social behaviors, while both critical theory and social conflict theory highlight AI's relationship with deepening economic and social gaps. Social structural theory emphasizes the importance of balancing technological development and social cohesion, while postmodern theory reflects the complexities of identity and truth in the digital age. Conclusions: The study suggested conducting research on the effects of artificial intelligence on (relationships, identity, values, and behaviors) in light of social theories and based on the foundations and methodologies of sociology.

Artificial Intelligence has recently become synonymous with any technology, be it in any field; the paradigm shift is being observed in every sector, and the ongoing discourse is being dominated by this talk. With its growing influence on individuals and, as a whole, on society, it is imperative to reflect on its outcomes. The growing integration of artificial intelligence (AI) in the education sector has been transforming the traditional learning and socialization process, and it is also presenting concerns. This paper dives into a primary understanding of AI technology, specifically its influence on the development of young students in terms of social interactions and cultural dynamics. This paper studied the AI’s influence on socialization and cultural dynamics through the lens of Vygotsky’s sociocultural theory and Bandura’s social learning theory. While the analysis reveals that AI does enhance personal learning and global collaborations, there is a risk of a reduction in face-to-face interactions, which could weaken skills such as empathy and nonverbal communication. Further, the finding also suggests that AI facilitates cross-cultural exposure, but there is a possible risk of cultural homogenization, which could marginalize the homegrown traditions. Data privacy, algorithmic bias, and social isolation are the other areas of concern that would further complicate AI’s role in education.

What entities deserve moral rights? Although this question has been of long-standing interest to ethics and social theory, it has resurfaced in a particularly pronounced way with artificial intelligence (AI). Using original survey data from a convenience sample and a large-scale random sample of Switzerland, we compare two explanations for the willingness to grant moral rights to AI. The ontological view refers to perceived human-like properties of AI, such as sentience (i.e., what AI is). The social impact view refers to the social-relational meaning of AI in terms of its impact on humans and trust (i.e., what AI does). The findings corroborate both views. Support for the general idea of AI rights and a political referendum granting basic rights to AI are positively correlated with the perceived ontological proximity of AI to humans, expected benefits of AI for society, and trust in AI developers. A right to life and physical integrity rests on perceived ontological properties only. Ontological distance of AI to simple computer programs, expected costs of AI, and trust in the state play a minor or no role. Sociological perspectives on an entity’s social-relational meaning are hence indispensable to explain the social construction of moral rights for AI.

This in-depth research explores the emerging relationship between artificial intelligence (AI) and legal systems by addressing key questions and understanding the evolution of global justice systems. This study focuses on the role of AI in strengthening the efficiency and objectivity of the judiciary, especially through the application of AI as judges in countries such as China and Estonia. This research aims to systematically analyse these developments, examining how AI is being integrated into justice systems in different parts of the world with challenges related to ethics, accountability, and human rights. The study results show that the integration of AI in the legal system brings increased efficiency and potential for transparency but also raises serious concerns about bias in AI algorithms, limitations in interpreting complex laws, and the impact on human rights principles. The main findings of this research show that the integration of AI in the legal system contains great potential for transformation but also requires a careful approach. While AI can improve the efficiency and quality of decision-making, it is important that AI is developed and implemented within a solid legal and ethical framework that respects human rights and maintains the justice system's integrity. This research emphasizes the need to consider each country's unique legal, cultural, and social context when adopting AI into their legal systems.

The thriving contemporary form of artificial intelligence (AI) called machine learning is often represented sensationally in popular media as a semi-mystical technology. Machine learning systems are frequently ascribed anthropomorphic capacities for learning, emoting and reasoning which, it is suggested, might lead to the alleviation of humanity’s woes. One critical reaction to such sensational proclamations has been to focus on the mundane reality of contemporary machine learning as mere inductive prediction based on statistical generalizations, albeit with surprisingly powerful abilities (Pasquinelli 2017). While the deflationist reaction is a necessary reply to sensationalist agitation, adequate comprehension of modern AI cannot be achieved while neglecting its material and social context. One does not have to subscribe wholeheartedly to the social construction of technology thesis1 to allow that the development and evolution of technologies are influenced by social factors. For AI, the most important aspect of the current social context is arguably capital, which increasingly dominates AI research and production. One former computer science professor describes a “giant sucking sound of [AI] academics going into industry” (Metz 2017). This paper introduces capital’s theory of AI as utility and initiates a discussion on its social consequences. First, I discuss utilities and their infrastructures and introduce a few critical thoughts on the topic. Second, I situate modern AI by way of a brief history. Third, I detail capital’s view of AI as a utility and the technical details underpinning it. Fourth, I sketch how AI as a utility frames a social problematic beyond the important issues of algorithmic bias and the automation of work. I do so by extrapolating from one consequence of AI as a utility which multiple capitalist firms predict: the curation of human subjectivities.

Schizophrenia researchers have focused on phenomenological aspects of the disorder to better understand its underlying nature. In particular, development of personal narratives-that is, the complexity with which people form, organize, and articulate their "life stories"-has recently been investigated in individuals with schizophrenia. However, less is known about how aspects of narrative relate to indicators of neurocognitive and social functioning. The objective of the present study was to investigate the association of linguistic complexity of life-story narratives to measures of cognitive and social problem-solving abilities among people with schizophrenia. Thirty-two individuals with a diagnosis of schizophrenia completed a research battery consisting of clinical interviews, a life-story narrative, neurocognitive testing, and a measure assessing multiple aspects of social problem solving. Narrative interviews were assessed for linguistic complexity using computerized technology. The results indicate differential relationships of linguistic complexity and neurocognition to domains of social problem-solving skills. More specifically, although neurocognition predicted how well one could both describe and enact a solution to a social problem, linguistic complexity alone was associated with accurately recognizing that a social problem had occurred. In addition, linguistic complexity appears to be a cognitive factor that is discernible from other broader measures of neurocognition. Linguistic complexity may be more relevant in understanding earlier steps of the social problem-solving process than more traditional, broad measures of cognition, and thus is relevant in conceptualizing treatment targets. These findings also support the relevance of developing narrative-focused psychotherapies. (PsycINFO Database Record

Aim/Purpose: This paper aims to investigate and understand the intentions of management undergraduate students in Hangzhou, China, regarding the adoption of Artificial Intelligence (AI) technologies in their education. It addresses the need to explore the factors influencing AI adoption in the educational context and contribute to the ongoing discourse on technology integration in higher education. Background: The paper addresses the problem by conducting a comprehensive investigation into the perceptions of management undergraduate students in Hangzhou, China, regarding the adoption of AI in education. The study explores various factors, including Perceived Relative Advantage and Trialability, to shed light on the nuanced dynamics influencing AI technology adoption in the context of higher education. Methodology: The study employs a quantitative research approach, utilizing the Confirmatory Tetrad Analysis (CTA) and Partial Least Squares Structural Equation Modeling (PLS-SEM) methodologies. The research sample consists of management undergraduate students in Hangzhou, China, and the methods include data screening, principal component analysis, confirmatory tetrad analysis, and evaluation of the measurement and structural models. We used a random sampling method to distribute 420 online, self-administered questionnaires among management students aged 18 to 21 at universities in Hangzhou. Contribution: This paper explores how management students in Hangzhou, China, perceive the adoption of AI in education. It identifies factors that influence AI adoption intention. Furthermore, the study emphasizes the complex nature of technology adoption in the changing educational technology landscape. It offers a thorough comprehension of this process while challenging and expanding the existing literature by revealing the insignificant impacts of certain factors. This highlights the need for an approach to AI integration in education that is context-specific and culturally sensitive. Findings: The study highlights students’ positive attitudes toward integrating AI in educational settings. Perceived relative advantage and trialability were found to impact AI adoption intention significantly. AI adoption is influenced by social and cultural contexts rather than factors like compatibility, complexity, and observability. Peer influence, instructor guidance, and the university environment were identified as pivotal in shaping students’ attitudes toward AI technologies. Recommendations for Practitioners: To promote the use of AI among management students in Hangzhou, practitioners should highlight the benefits and the ease of testing these technologies. It is essential to create communication strategies tailored to the student’s needs, consider cultural differences, and utilize the influence of peers and instructors. Establishing a supportive environment within the university that encourages innovation through policies and regulations is vital. Additionally, it is recommended that students’ attitudes towards AI be monitored constantly, and strategies adjusted accordingly to keep up with the changing technological landscape. Recommendation for Researchers: Researchers should conduct cross-disciplinary and cross-cultural studies with qualitative and longitudinal research designs to understand factors affecting AI adoption in education. It is essential to investigate compatibility, complexity, observability, individual attitudes, prior experience, and the evolving role of peers and instructors. Impact on Society: The study’s insights into the positive attitudes of management students in Hangzhou, China, toward AI adoption in education have broader societal implications. It reflects a readiness for transformative educational experiences in a region known for technological advancements. However, the study also underscores the importance of cautious integration, considering associated risks like data privacy and biases to ensure equitable benefits and uphold educational values. Future Research: Future research should delve into AI adoption in various academic disciplines and regions, employing longitudinal designs and qualitative methods to understand cultural influences and the roles of peers and instructors. Investigating moderating factors influencing specific factors’ relationship with AI adoption intention is essential for a comprehensive understanding.

XAI (eXplainable Artificial Intelligence) is an artificial intelligence technology that analyzes the causal relationship of artificial intelligence decision-making, finds appropriate evidence, and explains decision-making results at the user level. Customers of the XAI model are practitioners of financial institutions that operate the model and provide results to customers, and financial consumers who use the results of the AI algorithm model. The financial industry generates profits by managing individual and institutional funds, but also has responsibility for its management. Although artificial intelligence algorithms are being applied to tasks such as the discovery of financial crimes and cyber-crime that are becoming more intelligent and repetitive credit evaluations, unexpected algorithm vulnerabilities can cause losses due to incorrect judgments. XAI presents the basis for the results of artificial intelligence algorithms in the financial sector, and financial companies to which XAI is applied can reduce the possibility of wrong decision-making by artificial intelligence. This study aims to explore the XAI characteristics and influencing factors in the financial sector that affect the reliability and satisfaction of domestic financial consumers, and examine how these factors affect innovation performance in the financial sector. DARPA mentioned user satisfaction, explanation model level, task performance improvement, reliability evaluation, and error correction level as XAI's competency evaluation factors (competency evaluation index). This study refers to DARPA's XAI competency evaluation index, G20's human-centered artificial intelligence society principles, and the Financial Services Commission's financial sector artificial intelligence guidelines, etc. In addition, the innovation performance of financial institutions according to the introduction of XAI technology and service supply was set as a dependent variable. In order to verify the research model, an online survey was conducted targeting workers in companies introducing XAI in the financial sector. As a result of the study, it was found that the XAI characteristics of the financial sector (transparency, explainability, reliability, and result demonstrability) had a significant effect on innovation performance in the financial sector.

Introduction. The article analyzes the semiotic aspects of the representation of artificial intelligence in the socio-cultural space and their influence on the nature of social development. The spread and penetration of technologies into social and political processes using artificial intelligence (AI) is accompanied by active discussions in the socio-political environment. Artificial intelligence is considered as one of the most important resources of social and political development, so the desire of states, politicians and public organizations to determine the semantic framework of its public perception is understandable.Methodology and sources. The theoretical foundations of the work were the classic works on semiotics by Ch. S. Pierce, Ch. W. Morris, as well as theoretical works on modern political semiotics (I.V. Fomin, M.V. Ilyin and others). To work with the empirical base, analyzes of documents (reports and materials from UNESCO, the European Commission, the Royal Society of London, Stanford University), public opinion polls (VTsIOM), discourse analysis, semantic and content analysis of materials from Russian and foreign media were used.Results and discussion. As a result of the study, we found that the main contradiction that accompanies the understanding of artificial intelligence, its essence and purpose at all levels of socio-political discourse is the polarization of public ideas about AI as a control tool and (or) a development tool. The contextual fields were also outlined, within which further development of the semiosis of artificial intelligence is possible.Conclusion. To understand the nature of the representation of technologies using AI in socio-political discourse and to manage this process, an interdisciplinary approach and an appeal to the scientific potential of the social sciences and the humanities, in particular, semiotics, are appropriate, which makes it possible to study the semiosis of technology at the semantic, syntactic and pragmatic levels. It also seems promising to turn to a comparative analysis of cultural and historical traditions and national-state models of social management that specifically integrate AI into social processes.

The variability in responses generated by Large Language Models (LLMs) like OpenAI’s GPT-4 poses challenges in ensuring consistent accuracy on medical knowledge assessments, such as the United States Medical Licensing Exam (USMLE). This study introduces a novel multi-agent framework—referred to as a "Council of AIs"—to enhance LLM performance through collaborative decision-making. The Council consists of multiple GPT-4 instances that iteratively discuss and reach consensus on answers facilitated by a designated "Facilitator AI." This methodology was applied to 325 USMLE questions across Step 1, Step 2 Clinical Knowledge (CK), and Step 3 exams. The Council achieved consensus responses that were correct 97%, 93%, and 94% of the time for Step 1, Step 2CK, and Step 3, respectively, outperforming single-instance GPT-4 models. In cases where there wasn’t an initial unanimous response, the Council of AI deliberations achieved a consensus that was the correct answer 83% of the time. For questions that required deliberation, the Council corrected over half (53%) of responses that majority vote had gotten incorrect. At the end of deliberation, the Council often corrected majority responses that were initially incorrect: the odds of a majority voting response changing from incorrect to correct were 5 (95% CI: 1.1, 22.8) times higher than the odds of changing from correct to incorrect after discussion. We additionally characterized the semantic entropy of the response space for each question and found that deliberations impact entropy of the response space and steadily decrease it, consistently reaching an entropy of zero in all instances. This study showed that in a Council model response variability—often viewed as a limitation—could be leveraged as a strength, enabling adaptive reasoning and collaborative refinement of answers. These findings suggest new paradigms for AI implementation and reveal diversity of responses as a strength in collective decision-making even in medical question scenarios where there is a single correct response.Author SummaryIn our study, we explored how collaboration among multiple artificial intelligence (AI) systems could improve accuracy on medical licensing exams. While individual AI models like GPT-4 often produce varying answers to the same question—a challenge known as "response variability"—we designed a "Council of AIs" to turn this variability into a strength. The Council consists of several AI models working together, discussing their answers through an iterative process until they reach consensus.When tested on 325 medical exam questions, the Council achieved 97%, 93%, and 94% accuracy on the Step 1, Step 2CK, and Step 3, respectively. This improvement was most notable when answers required debate: in cases where initial responses disagreed, the collaborative process corrected errors 83% of the time. Our findings suggest that collective decision-making— even among AIs—can enhance accuracy and AI collaboration can potentially lead to more trustworthy tools for healthcare, where accuracy is critical. By demonstrating that diverse AI perspectives can refine answers, we challenge the notion that consistency alone defines a "good" AI. Instead, embracing variability through teamwork might unlock new possibilities for AI in medicine and beyond. This approach could inspire future systems where AIs and humans collaborate (e.g. on Councils with both humans and AIs), combining strengths to solve complex problems. While technical challenges remain, our work highlights a promising path toward more robust, adaptable AI solutions.