Phase animalism and artificial intelligence

In this final set of explorations/meditations (of three), we examine the requirements for developing artificial general intelligence (AGI) through the lens of human cognitive architecture, with particular emphasis on the role of narrative selfhood and social cognition. Drawing on perspectives from cognitive science, philosophy of mind, and artificial intelligence research, we critically evaluate current claims about the capabilities of large language models, particularly regarding their purported achievements of theory of mind and self-awareness. We argue that genuinely human-like artificial intelligence may require more than sophisticated pattern recognition and language modeling, potentially necessitating the development of coherent narrative self-models and rich causal understanding. Special attention is given to the relationship between consciousness, conscience, and trustworthy AI systems, suggesting that meaningful artificial intelligence may require forms of richly-embodied and socially-embedded development to achieve robust and reliable functionality. We conclude by proposing that the path to artificial general intelligence may require recapitulating aspects of human cognitive development, particularly regarding the construction of narrative identity and social-moral reasoning capabilities. This analysis has implications for both the technical development of AI systems and the ethical frameworks through which we evaluate artificial minds.

AI can even outperform humans in many tasks such as winning games like Go and poker as well as engaging in creative endeavours in writing novels and music. Despite this, it is still a long way from building artificial human intelligence. Current AIs are only designed to excel in their intended functions and cannot generate knowledge to new tasks and situations. For AI to achieve artificial human intelligence requires us to study and understand the human brain. Neuroscience is the study of the anatomy and physiology of the human brain. It provides us interesting insights into how the brain works to develop better AI systems. Conversely, better AI systems can help drive neuroscience forward and further unlock the secrets of the brain. Neuroscience and AI are closely related. The synergy of the two will benefit each other. Besides the benefits of neuroscience for AI research, neuroscience also has important implications for machine learning. This chapter discusses the implications of neuroscience for general artificial intelligence and the benefits of AI for neuroscience research.

In this article, we review a number of pros and cons, including the meaning and application of artificial intelligence. We also considered the latest technological developments in the real world and discussed the application of AI in the fields of heavy industry, gaming, aviation, weather forecasting and expert systems, focusing on expert systems. Specialty applications for AI include expert systems, speech recognition, and machine vision. AI can be classified as weak or strong. Weak AI (also known as Narrow AI) is an AI system designed and trained for specific tasks. Virtual personal assistants (such as Apple's Siri) are weak AI. Powerful AI, also known as artificial intelligence, is an AI system with a wide range of human cognitive capabilities. When performing unfamiliar tasks, powerful AI systems can find solutions without manual intervention. The artificial intelligence that makes computers behave like humans is human-like, fashion-like, and humans spend much less time. Therefore, it is called artificial intelligence. This article concludes by analyzing the future potential of artificial intelligence.

There is great interest in the application of ‘artificial intelligence’ (AI) to pharmacovigilance (PV). Although US FDA is broadly exploring the use of AI for PV, we focus on the application of AI to the processing and evaluation of Individual Case Safety Reports (ICSRs) submitted to the FDA Adverse Event Reporting System (FAERS). We describe a general framework for considering the readiness of AI for PV, followed by some examples of the application of AI to ICSR processing and evaluation in industry and FDA. We conclude that AI can usefully be applied to some aspects of ICSR processing and evaluation, but the performance of current AI algorithms requires a ‘human-in-the-loop’ to ensure good quality. We identify outstanding scientific and policy issues to be addressed before the full potential of AI can be exploited for ICSR processing and evaluation, including approaches to quality assurance of ‘human-in-the-loop’ AI systems, large-scale, publicly available training datasets, a well-defined and computable ‘cognitive framework’, a formal sociotechnical framework for applying AI to PV, and development of best practices for applying AI to PV. Practical experience with stepwise implementation of AI for ICSR processing and evaluation will likely provide important lessons that will inform the necessary policy and regulatory framework to facilitate widespread adoption and provide a foundation for further development of AI approaches to other aspects of PV.

This book offers a unique interdisciplinary perspective on the ethics of 'artificial intelligence' – autonomous, intelligent, (and connected) systems, or AISs, applying principles of social cognition to understand the social and ethical issues associated with the creation, adoption, and implementation of AISs. As humans become entangled in sociotechnical systems defined by human and artificial agents, there is a pressing need to understand how trust is created, used, and abused. Compounding the difficulty in answering these questions, stakeholders directly or indirectly affected by these systems differ in their motivations, understanding, and values. This volume provides a comprehensive resource to help stakeholders understand ethical issues of designing and implementing AISs using an ethical sensemaking approach. Starting with the general technical affordances of AIS, Dr. Jordan Richard Schoenherr considers the features of system design relating data integrity, selection and interpretation of algorithms, and the evolution processes that drive AISs innovation as a sociotechnological system. The poles of technophobia (algorithmic aversion) and technophilia (algorithmic preference) in the public perception of AISs are then described and considered against existing evidence, including issues ranging from the displacement and re-education needs of the human workforce, the impact of use of technology on interpersonal accord, and surveillance and cybersecurity. Ethical frameworks that provide tools for evaluating the values and outcomes of AISs are then reviewed, and how they can be aligned with ethical sensemaking processes identified by psychological science is explored. Finally, these disparate threads are brought together in a design framework. Also including sections on policies and guideline, gaming and social media, and Eastern philosophical frameworks, this is fascinating reading for students and academics in psychology, computer science, philosophy, and related areas, as well as professionals such as policy makers and those working with AI systems.

In modern conditions, the use of AI is an advantage for business. This allows you to free up additional human resources and direct them to other tasks, speed up operations, and move to new areas of development. At the same time, the use of AI leads to an increase in risks, cyber threats, costs, and the possibilities of minimizing risk using AI determine the relevance of this study. In this scientific work, the subject of research is the process of applying and using AI in the banking business and in the banking risk management. The topics specified for the study were applied standard scientific methods of analysis, synthesis, deduction, and induction, it is possible to determine the directions of using AI in the banking sector and identify the possibilities of using AI together with risk management tools. The main goal of the research is to identify the advantages and threats of using AI in the banking sector. Taking into account the clear regulation of banks' activities, the paper highlights risk management tools: creation and maintenance of a database of internal and external operational risk events, key operational risk indicators, operational risk self-assessment, scenario analysis, mathematical modelling, analysis of process maps, comparative analysis. Today, almost all bank operations can be performed based on AI or systems that use it. Examples of where such tools are used include financial monitoring, active or treasury operations, securitization, Chat-bots and social networks, remote identification and office management. At the same time, AI systems are also used to control the risks arising in these processes: for process control and risk management. Each of the above risk management tools can be based on AI systems, which opens up significant opportunities for control, risk minimization and the release of human resources. However, one should not forget about potential threats that can worsen the quality of risk management: lack of specialists, data fragmentation in banks, validation problems, the cost of connecting AI, imperfection of existing models, and the possibility of data loss. In conclusion, it is worth noting that the use of artificial intelligence, along with significant advantages for banks, generates numerous risks that can be minimized with the use of the same artificial intelligence. This situation indicates a transition to a new stage of banking risk management – the stage of cooperation with artificial intelligence.

As AI systems are integrated into high stakes social domains, researchers now examine how to design and operate them in a safe and ethical manner. However, the criteria for identifying and diagnosing safety risks in complex social contexts remain unclear and contested. In this paper, we examine the vagueness in debates about the safety and ethical behavior of AI systems. We show how this vagueness cannot be resolved through mathematical formalism alone, instead requiring deliberation about the politics of development as well as the context of deployment. Drawing from a new sociotechnical lexicon, we redefine vagueness in terms of distinct design challenges at key stages in AI system development. The resulting framework of Hard Choices in Artificial Intelligence (HCAI) empowers developers by 1) identifying points of overlap between design decisions and major sociotechnical challenges; 2) motivating the creation of stakeholder feedback channels so that safety issues can be exhaustively addressed. As such, HCAI contributes to a timely debate about the status of AI development in democratic societies, arguing that deliberation should be the goal of AI Safety, not just the procedure by which it is ensured.

AI systems are complex and difficult to understand, which makes it essential to establish a basis of trust between the user and the system to enable comfortable interactions. AI systems present both opportunities and risks, hence developers should endeavor to minimize risks as much as possible. Users, particularly in the realm of safety-critical systems, require sufficient trust in any new technology to consider its usage. Uncertainty and missing understanding about the field of AI and algorithms in general hinders and stops its usage. Therefore, transparency and comprehensibility should be goals in the development of AI systems. These factors can be divided into three groups: Technical factors, social factors, and evaluation factors. Each of these groups consists of five factors that can either enhance or diminish trust in an AI system. Based on these factors, a reference model is created to classify these three groups of factors within the overall context of trust formation. A study was conducted with 258 participants using a questionnaire to examine and subsequently analyze and evaluate the previously formulated reference model. An attempt is made to hierarchize the individual factors within the three groups and to identify any differences related to safety-critical and non-safety-critical systems. Keywords: artificial intelligence, trustworthy ai, responsible ai

The convergence of explainable artificial intelligence and robust AI systems represents one of the most critical challenges in contemporary machine learning research. This paper presents a unified theoretical framework that addresses the fundamental tension between model interpretability and adversarial robustness, two properties traditionally viewed as conflicting objectives in AI system design. Through a comprehensive analysis of the mathematical foundations underlying both explainability and robustness, we demonstrate that these characteristics can be synergistically integrated rather than traded off against each other. The proposed framework establishes theoretical connections between explanation quality metrics and adversarial vulnerability measures, revealing that well-explained models can actually exhibit enhanced robustness when properly constructed. Our approach introduces novel methodologies for simultaneously optimizing interpretability and security in AI systems, with particular emphasis on applications in critical domains where both transparency and reliability are essential. The framework provides practical guidelines for developing AI systems that maintain high performance while offering meaningful explanations and demonstrating resilience against adversarial attacks. This work contributes to the emerging field of trustworthy AI by providing both theoretical foundations and practical methodologies for building systems that are simultaneously explainable, robust, and reliable.

The field of AI safety aims to prevent increasingly capable artificially intelligent systems from causing humans harm. Research on moral alignment is widely thought to offer a promising safety strategy: if we can equip AI systems with appropriate ethical rules, according to this line of thought, they’ll be unlikely to disempower, destroy or otherwise seriously harm us. Deontological morality looks like a particularly attractive candidate for an alignment target, given its popularity, relative technical tractability and commitment to harm-avoidance principles. I argue that the connection between moral alignment and safe behavior is more tenuous than many have hoped. In general, AI systems can possess either of these properties in the absence of the other, and we should favor safety when the two conflict. In particular, advanced AI systems governed by standard versions of deontology need not be especially safe.

Theoretical background: The dynamic development of generative artificial intelligence such as ChatGPT has transformed the perception of creative work. In the graphic realm, AI systems like Midjourney, DALL-E, or Adobe Firefly allow the creation of high-quality graphics without the need for artistic skills or hiring a talented designer. Concurrently, the emergence of cryptocurrencies and the associated non-fungible tokens (NFTs) has resulted in radical changes in the creative sector, especially in the art market. Purpose of the article: The aim of this article is to examine the development pace and impact of AI-generated art and NFTs on the global art market, focusing on market trends, dependency on energy prices, segmentation, and how these changes influence artwork pricing and artists’ livelihoods. Research methods: The author has assessed the popularity and development of the NFT market, as well as the main reasons for its collapse in 2022. The author identified possible scenarios for the development of the art market, pointed out the primary potential threats, and highlighted the key determinants of future digital asset valuation. Main findings: Despite initial euphoria, buyers depreciate digital goods, especially those generated by artificial intelligence. They are considered inherently inferior and less valuable. Overproduction of works combined with the availability of AI solutions means that the traditional supply and demand mechanism lowers the price of assets and thereby forces more and more graphic designers, photographers, and painters to abandon their professions. Competition from artificial intelligence is subject to the same supply and demand mechanisms, which reduces the cost of access to quality graphics for entrepreneurs and individuals. Simultaneously, as the market becomes saturated with synthetic goods, there will be a delineation at the segment level, similar to what has happened with artisanal beers, hand-assembled cars, and furniture, or handmade ceramics. An unwritten, "made by humans" certificate will result in a 300–500% higher price for similar goods made by humans compared to works of artificial intelligence. This situation will resemble the division of the clothing or furniture market into mass-produced goods and designer items. Amid all this, the increasingly prominent role of NFTs will be evident, which will also appreciate in value, but due to the ecological taxation resulting from energy consumption that is 100,000 times higher than that of a regular bank payment.

It is not a coincidence that the research fields of Artificial Intelligence and of Law have met, and merged into the interdisciplinary research field of Artificial Intelligence and Law. Both fields use formal methods, with their strengths and limitations; in AI there are software, logic and statistics, in Law there are statutes, procedures and institutions. Both fields are creative; in AI systems are built, experiments designed and paradigms replaced, in Law regulations are passed by lawmakers, precedents are set and ideologies balanced. Both fields struggle with the inevitable complexity of modeling human behavior; in AI with the goal to reconstruct human behavior, in Law with the goal to steer human behavior. These and other similarities are driving the active and dedicated community of AI and Law. Researchers are taking their inspiration from the Law with its insights developed over millennia combining them with AI's half a century of lessons. The beginnings of AI and Law are marked by the first International Conference on Artificial Intelligence and Law in Boston in 1987, a quarter of a century ago. Ever since that first conference, the biennial ICAIL conference series has been a primary forum for the exchange and discussion of the latest research insights in the interdisciplinary field of Artificial Intelligence and Law. The Fourteenth International Conference on Artificial Intelligence and Law (Rome, Italy, June 10-14, 2013) continues from these 25 year old achievements, and provides a program that consists of invited lectures, full papers, research abstracts, system demonstrations, workshops and tutorials.

The “Basic Act on the Promotion of Foundations for Development and Trust for Artificial Intelligence” (hereinafter referred to as the “Basic Act on Artificial Intelligence”), enacted on January 21, 2025, marks South Korea as the second country in the world, following the European Union, to establish a comprehensive legal framework for artificial intelligence. This legislation is significant in that it provides a legal foundation for both the advancement of AI technology and its ethical utilization. However, for the Basic Act on Artificial Intelligence to be truly effective and responsive to the rapidly evolving technological landscape and its diverse social impacts, more sophisticated and balanced legal refinements are required. First, a differentiated regulatory system based on the risk level of AI systems must be established. By harmoniously combining the EU’s systematic risk classification and comprehensive regulatory approach with the United States’ flexible legal application, Korea should build a balanced regulatory framework that simultaneously fosters technological innovation, ensures ethical responsibility, and protects citizens’ rights. Second, transparency and explainability of AI must be enhanced. This includes clearly stipulating concrete methods for explaining AI decision-making processes and outcomes, procedures for objections, obligations for detailed technical documentation and preservation, and requirements for logging and record-keeping, thereby securing public trust in AI. Third, the protection of ethics and human rights in AI must be strengthened. This can be achieved by mandating impact assessments for high-impact AI, introducing mechanisms to prevent discrimination by AI systems, and ensuring meaningful human intervention. Fourth, to balance AI industry development with regulation, Korea should consider adopting a risk-based approach, flexible application of existing legal systems, building innovation-supporting infrastructure, strengthening international cooperation, and introducing AI-specific regulatory sandboxes. Through such improvements, the Basic Act on Artificial Intelligence can achieve a balanced realization of technological innovation, ethical use of AI, and protection of citizens’ rights. It is essential that these elements are thoroughly reflected in the forthcoming enforcement decrees, and that continuous monitoring and improvement are pursued to enhance the effectiveness of AI governance in Korea.

Across the sciences, each discipline develops its own domain ontology—its structural assumptions about what exists, from particles and fields to molecules, cells, minds, and social agents. Yet beneath this disciplinary diversity lies a deeper unity. This paper presents an empirically grounded structural ontology showing that the universe organizes itself through four Simple Interaction Rules (SIRs): Distinctions, Systems, Relationships, and Perspectives (DSRP). Using the Unified Calculus of Organization, we show that these operators are not independent categories but projections of one generative polarity expressed through eight elemental roles and three universal dynamics (4–8–3). These structures appear empirically in quantum state boundaries, molecular bonding symmetries, cellular architectures, ecological networks, social systems, and human cognition. They constitute the structural grammar of reality. Two lines of scientific evidence support this claim. First, existence evidence demonstrates that the DSRP invariants—polarity, part–whole recursion, relational interaction, and perspectival orientation—are instantiated in nature independent of human observers. Second, effect evidence shows that human cognition uses the same invariants but with measurable structural bias. When these biases are corrected, individuals exhibit large and replicable gains in systems thinking, metacognition, causal reasoning, and model accuracy. In an age dominated by large language models, this structural ontology becomes more than a philosophical insight—it becomes a practical necessity. Current LLM-based AI systems lack internal ontology: they do not represent boundaries, part–whole structure, causal or conceptual relationships, or perspectives. As a result, they hallucinate, contradict themselves, collapse distinctions, fabricate causal chains, and drift semantically over time. Without structure, there is no coherence. By formalizing the generative grammar of reality and cognition, the ontology advanced here provides the missing structural foundation that both humans and AI systems require to think, model, and understand reliably. It clarifies why humans reason the way they do, why AI systems fail the way they do, and why a shared structural framework—one that makes thinking visible, measurable, and governable—is essential for science, education, and human–AI intelligence. DSRP is not a theory of systems; it is the structural ontology that makes systems possible—and the generative grammar future human and artificial intelligence will require.

AI solutions are frequently presented as promising solutions to a wide range of global challenges, including public health. However, the potential they hold for advancing public health and the challenges to their deployment require examination particularly in the global south. Using a narrative synthesis on several recent reports of AI solutions in public health and their challenges, we explain the opportunities and challenges of using AI to address major public health problems, particularly in the global south, and describe the methods and models employed by AI systems. In addition to surveillance and monitoring, AI systems have been deployed for disease screening, early diagnosis and management, health education, epidemic forecasting, and for health program planning and management for a range of public health issues. Challenges persist in the interpretability, explainability, generalizability and feasibility of these systems. Also, regulatory, legal, and ethical oversight have not kept pace with AI development. More significantly, existing gaps in health system infrastructure, information system quality, and workforce capacity impede real world implementation of AI for public health. We explain each of these issues and the challenges they pose to using AI in public health. We recommend more research into real-world implementations, and the addressal of persistent health system constraints, to realise the transformative potential AI holds for public health.