Since the release of OpenAI's ChatGPT, the world has entered a race to develop more capable and powerful AI, including artificial general intelligence (AGI). The development is constrained by the dependency of AI on the model, quality, and quantity of training data, making the AI training process highly costly in terms of resources and environmental consequences. Thus, improving the effectiveness and efficiency of the AI training process is essential, especially when the Earth is approaching the climate tipping (...) points and planetary boundaries. It is evident that AI can function better when trained to mimic certain human mental processes. Based on insights from quantum mechanics and the informational entropy-based notion of value, we suggest AI developers can enhance the AI training processes' effectiveness and efficiency by creating and implementing parameter systems that are capable of effectively assigning probabilities to informational quanta. Such systems can help reduce entropy within the system, which lowers the energy required for data storage and processing while decreasing the likelihood of information loss during training. Successfully applying the concept of information entropy-based value in AI development will be crucial for advancing generative AI and achieving AGI in a sustainable manner. (shrink)

The influence of AI has spread globally, intriguing both the East and the West. As a result, some Chinese scholars have explored how AI and Chinese philosophy can be examined together, and have offered some unique insights into AI from a Chinese philosophical perspective. Similarly, we investigate how the two fields can be developed in conjunction, focusing on the popular Confucian philosophy. In this work, we use Confucianism as a philosophical foundation to investigate human-technology relations closely, proposing that a Confucian-imbued (...) AI would reduce prevalent AI ethical issues. Through this discussion, the strong substantivist stance that we will lose control over AI is refuted. We then discuss the practical elements of this project, explaining how passages from the Confucian text of the Analects can be incorporated into AI systems. As an example of a real-world context, psychology is used to demonstrate how the project could be practically applied. Potential criticisms surrounding the proposed project are also identified and challenged. By reducing the negative effects of common AI ethical issues and utilising Confucian principles from The Analects, AI systems can be more ethical, promoting peaceful relations between humans and technology. (shrink)

AI technology promises to be both the most socially important and the most profitable technology of a generation. At the same time, the control over – and profits from – the technology is highly concentrated to a handful of large tech companies. This chapter discusses whether bringing AI technology under collective ownership and control is an attractive way of counteracting this development. It discusses justice-based rationales for collective ownership, such as the claim that, since the training of AI systems (...) relies on a form of enclosing of the data commons, the value created by those systems should be fairly distributed. It also considers democracy-based rationales, like the suggestion that collective ownership is needed to ensure democratic control over the way this crucial technology is being developed and deployed. The paper also discusses possible forms of collective ownership, like publicly funded advanced AI research and democratically controlled AI companies. It concludes that the case for shifting to a model with collective ownership is the most compelling when based on the concern that the private model could come to concentrate economic and political power to such a degree that it threatens institutions designed to promote justice and democracy. (shrink)

The increasing demand for transparency in AI has recently come under scrutiny. The question is often posted in terms of “epistemic double standards”, and whether the standards for transparency in AI ought to be higher than, or equivalent to, our standards for ordinary human reasoners. I agree that the push for increased transparency in AI deserves closer examination, and that comparing these standards to our standards of transparency for other opaque systems is an appropriate starting point. I suggest that a (...) more fruitful exploration of this question will involve a different comparison class. We routinely treat judgments made by highly trained experts in specialized fields as fair or well-grounded even though—by the nature of expert/layperson division of epistemic labor—an expert will not be able to provide an explanation of the reasoning behind these judgments that makes sense to most other people. Regardless, laypeople are thought to be acting reasonably—and ethically—in deferring to the judgments of experts that concern their areas of specialization. I suggest that we reframe our question regarding the appropriate standards of transparency in AI as one that asks when, why, and to what degree it would be ethical to accept opacity in AI. I argue that our epistemic relation to certain opaque AI technology may be relevantly similar to the layperson’s epistemic relation to the expert in certain respects, such that the successful expert/layperson division of epistemic labor can serve as a blueprint for the ethical use of opaque AI. (shrink)

This paper delves into the significance and challenges of Artificial Intelligence (AI) ethics and justice in terms of Common Good and Well-being, fairness and non-discrimination, rational public deliberation, and autonomy and control. Initially, the paper establishes the groundwork for subsequent discussions using the Academia Sinica LLM incident and the AI Technology R&D Guidelines of the National Science and Technology Council(NSTC) as a starting point. In terms of justice and ethics in AI, this research investigates whether AI can fulfill human common (...) interests and welfare. Taking AI injustice as an example, I analyze the practical assessment of AI regarding regional, industrial, and social impacts. Further, this paper discusses the challenges of fairness and non-discrimination in AI, specifically addressing the issue of training on biased data, discussing the acquisition of bias by AI and post-processing supervision issues, and emphasizing the importance of rational public deliberation in this process. Then, this research examines the challenges and countermeasures the rational public faces in public deliberation, such as the importance of education in STEM scientific literacy and technological capabilities. Finally, in discussing AI and autonomy, I propose a 'Human-Centered Approach’ rather than relying solely on the 'Technological Utility Maximization' brought by AI to achieve substantial AI justice. -/- Keywords: AI Ethics and Justice, Fairness and Non-Discrimination, Biased Data Training, Public Deliberation, Autonomy, Human-Centered Approach . (shrink)

Two art exhibitions, “Training Humans” and “Making Faces,” and the accompanying essay “Excavating AI: The politics of images in machine learning training sets” by Kate Crawford and Trevor Paglen, are making substantial impact on discourse taking place in the social and mass media networks, and some scholarly circles. Critical scrutiny reveals, however, a self-contradictory stance regarding informed consent for the use of facial images, as well as serious flaws in their critique of ML training sets. Our analysis (...) underlines the non-negotiability of informed consent when using human data in artistic and other contexts, and clarifies issues relating to the description of ML training sets. (shrink)

Recent work on interpretability in machine learning and AI has focused on the building of simplified models that approximate the true criteria used to make decisions. These models are a useful pedagogical device for teaching trained professionals how to predict what decisions will be made by the complex system, and most importantly how the system might break. However, when considering any such model it’s important to remember Box’s maxim that "All models are wrong but some are useful." We focus on (...) the distinction between these models and explanations in philosophy and sociology. These models can be understood as a "do it yourself kit" for explanations, allowing a practitioner to directly answer "what if questions" or generate contrastive explanations without external assistance. Although a valuable ability, giving these models as explanations appears more difficult than necessary, and other forms of explanation may not have the same trade-offs. We contrast the different schools of thought on what makes an explanation, and suggest that machine learning might benefit from viewing the problem more broadly. (shrink)

A key assumption fuelling optimism about the progress of Large Language Models (LLMs) in modelling the world is that the truth is systematic: true statements about the world form a whole that is not just consistent, in that it contains no contradictions, but cohesive, in that the truths are inferentially interlinked. This holds out the prospect that LLMs might rely on that systematicity to fill in gaps and correct inaccuracies in the training data: consistency and cohesiveness promise to facilitate (...) progress towards comprehensiveness in an LLM’s representation of the world. However, philosophers have identified reasons to doubt that the truth is systematic across all domains of thought, arguing that in normative domains, in particular, the truth is not necessarily systematic. I argue that insofar as the truth in normative domains is asystematic, this renders it correspondingly harder for LLMs to make progress, because they cannot rely on the consistency and cohesiveness of the truth to work towards comprehensiveness. And the less LLMs can rely on the systematicity of truth, the less we can rely on them to do our practical deliberation for us, as there is correspondingly more of a role for human agency in navigating asystematic normative domains. (shrink)

In artificial intelligence (AI), responses generated by machine-learning models (most often large language models) may be unfactual information presented as a fact. For example, a chatbot might state that the Mona Lisa was painted in 1815. Such phenomenon is called AI hallucinations, seeking inspiration from human psychology, with a great difference of AI ones being connected to unjustified beliefs (that is, AI “beliefs”) rather than perceptual failures). -/- AI hallucinations may have their source in the data itself, that is, the (...) source content, or in the training procedure, i.e. the way the knowledge was encoded in the model’s parameters, so that errors in encoding and decoding textual and non-textual representations can cause hallucinations. In this paper, we will observe how such errors come to life and how they might be mitigated. For this purpose, we will analyze the usability of justification logics, to behave as a proof checker for validating the correctness of large language models’ (LLM) responses. Justification logic was developed by S. Artemov, and later on mostly by Artemov and M. Fitting, deriving its main idea from the logic of proofs (LP): knowledge and belief modalities are seen as justification terms, i.e. t:X stands for t is a (proper) justification for X. Justification logic originated from attempts to create semantics for intuitionistic logic where proofs were the most proper justifications, but in further development, justification logic could be applied to different kinds of justifications). With the recent attempts to mitigate incorrect LLM responses, we will analyze various guardrails that are currently used for LLM responses, and see how the logic of justification may provide its benefits as an AI safety layer against false data. (shrink)

We investigate the application of Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) networks, and a hybrid CNN-LSTM model for forecasting air pollution levels based on historical data. Our experimental setup uses real-world air quality datasets from multiple regions, containing measurements of pollutants like PM2.5, PM10, CO, NO2, and SO2, alongside meteorological data such as temperature, humidity, and wind speed. The models are trained, validated, and tested using a split dataset, and their accuracy is evaluated using performance metrics like Mean (...) Absolute Error (MAE), Root Mean Squared Error (RMSE), and R-squared. (shrink)

This thesis focuses on the use of intelligent tutoring system for education and training of students specialized in electricity in the field of technical and vocational education. The use of modern systems in training and education will have a great positive impact in improving the level of students receiving training and education; this will improve the level of the local economy by producing students of professionals who are able to engage in society efficiently, especially for those who (...) have specialized in electricity. It is known that students in electrical specialties face many problems because of time and cost. Learning and training is not as easy as you imagine, so intelligence is how to overcome these limitations effectively. Motivating the Intelligent Tutorial System (ITS) for this type of training. This system consists of a student model, which in turn will contain lessons, exercises and examples that will be used by the student in addition to the user interface model which makes the teacher to be able to manage lessons, examples, and questions and make changes to the system. Expert System (ES) model representing the Artificial Intelligence (AI) here has been used to determine the level of the student and gives him the results, directions and levels. The program was implemented and was used by trainees to show the extent of its impact. The students expressed their surprise and admiration in this way of education as the system used focuses on individual education which can be used at any time and without restrictions. After this experience, the dedicated (ITS) system needs to be developed in several ways such as to be used on mobile systems and web systems in addition to linking some of the specialized programs in the field of electricity. (shrink)

Twenty-five years ago, my colleagues Miyuki Kamachi and Jiro Gyoba and I designed and photographed JAFFE, a set of facial expression images intended for use in a study of face perception. In 2019, without seeking permission or informing us, Kate Crawford and Trevor Paglen exhibited JAFFE in two widely publicized art shows. In addition, they published a nonfactual account of the images in the essay “Excavating AI: The Politics of Images in Machine Learning Training Sets.” The present article recounts (...) the creation of the JAFFE dataset and unravels each of Crawford and Paglen’s fallacious statements. I also discuss JAFFE more broadly in connection with research on facial expression, affective computing, and human-computer interaction. (shrink)

This paper reframes the issue of appropriation, extraction, and dispossession through AI—an assemblage of machine learning models trained on big data—in terms of enclosure and foreclosure. While enclosures are the product of a well-studied set of operations pertaining to both the constitution of the sovereign State and the primitive accumulation of capital, here, I want to recover an older form of the enclosure operation to then contrast it with foreclosure to better understand the effects of current algorithmic rationality. I argue (...) that the act of enclosing is to be understood as a set of fundamental operations that consist in producing structural distinctions between inside and outside, inclusion and exclusion—whether by drawing lines on a map, constructing border walls, or by algorithmically categorizing and (mis)recognizing people and things. Tracking the transformation of an enclosure-logic into one of foreclosure, I show how current AI perpetuates and at the same time expends forms of extraction, exclusion, and dispossession toward a totalizing horizon. For, while the outside is essential and constitutive to the logic of enclosure, foreclosure, by contrast, is characterized by the “totalizing desire” to not leave anything out of it. The totalizing desire to move beyond the logic of the enclosure forecloses that the enclosure has just gotten bigger and that the operations of exclusion have been masked and displaced, making them harder to contest. (shrink)

I conducted an experiment by using four different artificial intelligence models developed by OpenAI to estimate the persuasiveness and rational justification of various philosophical stances. The AI models used were text-davinci-003, text-ada-001, text-curie-001, and text-babbage-001, which differed in complexity and the size of their training data sets. For the philosophical stances, the list of 30 questions created by Bourget & Chalmers (2014) was used. The results indicate that it seems that each model has its own plausible ‘cognitive’ style. The (...) outcomes of the ‘strongest’ model correlate with the average philosophers' stance. (shrink)

Results indicate substantial improvements in efficiency, accuracy, and personalized care, but also highlight the challenges of data privacy, ethical considerations, and system scalability. By providing a structured analysis, this research contributes insights into optimizing generative AI deployments for both banking and healthcare, ensuring a balance between innovation and risk management. The study concludes with recommendations for future research directions, including advanced model training, ethical guidelines, and enhanced privacy measures. These insights aim to inform practitioners on the benefits of generative (...) AI, ensuring sustainable integration into banking and healthcare ecosystems. (shrink)

Just as humans can draw conclusions responsibly or irresponsibly, so too can computers. Machine learning systems that have been trained on data sets that include irresponsible judgments are likely to yield irresponsible predictions as outputs. In this paper I focus on a particular kind of inference a computer system might make: identification of the intentions with which a person acted on the basis of photographic evidence. Such inferences are liable to be morally objectionable, because of a way in which they (...) are presumptuous. After elaborating this moral concern, I explore the possibility that carefully procuring the training data for image recognition systems could ensure that the systems avoid the problem. The lesson of this paper extends beyond just the particular case of image recognition systems and the challenge of responsibly identifying a person’s intentions. Reflection on this particular case demonstrates the importance (as well as the difficulty) of evaluating machine learning systems and their training data from the standpoint of moral considerations that are not encompassed by ordinary assessments of predictive accuracy. (shrink)

The editorial introduces issue 21.1 of Technoetic Arts via a critical reflection on the artificial intelligence hype (AI hype) that emerged in 2022. Tracing the history of the critique of Large Language Models, the editorial underscores that there are substantial ethical challenges related to bias in the training data, copyright issues, as well as ecological challenges which the technology industry has consistently downplayed over the years. -/- The editorial highlights the distinction between the current AI technology’s reliance on extensive (...) pre-existing human-generated data for pattern recognition and the performative process inherent in art practice. Patterns of creative output can be replicated without a thorough understanding of underlying concepts. In contrast, the performative process of art practice, as explored through the conceptual frameworks of creativity by Noam Chomsky, Charles Sanders Peirce, and Roger T. Ames and discussed in previous publications by this editorial's author, creates concepts and continually ‘strives to extend itself towards the unknown.’ -/- Following this introductory section, the editorial provides an overview of the ten articles featured in this issue of Technoetic Arts. The first four texts engage with emerging technologies from various perspectives in art and design. The subsequent six articles, part of a special section titled 'Perspectives from Chandigarh,' reflect a quest for meaningful existence within colonial, patriarchal, and biopolitical structures that frame everyday practices of exclusion and oppression. (shrink)

This paper reframes the issue of appropriation, extraction, and dispossession through AI—an assemblage of machine learning models trained on big data—in terms of enclosure and foreclosure. While enclosures are the product of a well-studied set of operations pertaining to both the constitution of the sovereign State and the primitive accumulation of capital, here, I want to recover an older form of the enclosure operation to then contrast it with foreclosure to better understand the effects of current algorithmic rationality. I argue (...) that the act of enclosing is to be understood as a set of fundamental operations that consist in producing structural distinctions between inside and outside, inclusion and exclusion—whether by drawing lines on a map, constructing border walls, or by algorithmically categorizing and (mis)recognizing people and things. Tracking the transformation of an enclosure-logic into one of foreclosure, I show how current AI perpetuates and at the same time expends forms of extraction, exclusion, and dispossession toward a totalizing horizon. For, while the outside is essential and constitutive to the logic of enclosure, foreclosure, by contrast, is characterized by the “totalizing desire” to not leave anything out of it. The totalizing desire to move beyond the logic of the enclosure forecloses that the enclosure has just gotten bigger and that the operations of exclusion have been masked and displaced, making them harder to contest. (shrink)

The recent wave of generative AI (GAI) systems like Stable Diffusion that can produce images from human prompts raises controversial issues about creatorship, originality, creativity and copyright. This paper focuses on creatorship: who creates and should be credited with the outputs made with the help of GAI? Existing views on creatorship are mixed: some insist that GAI systems are mere tools, and human prompters are creators proper; others are more open to acknowledging more significant roles for GAI, but most conceive (...) of creatorship in an all-or-nothing fashion. We develop a novel view, called CCC (collective-centered creation), that improves on these existing positions. On CCC, GAI outputs are created by collectives in the first instance. Claims to creatorship come in degrees and depend on the nature and significance of individual contributions made by the various agents and entities involved, including users, GAI systems, developers, producers of training data and others. Importantly, CCC maintains that GAI systems can sometimes be part of a co-creating collective. We detail how CCC can advance existing debates and resolve controversies around creatorship involving GAI. (shrink)

Sora from Open AI has shown exceptional performance, yet it faces scrutiny over whether its technological prowess equates to an authentic comprehension of reality. Critics contend that it lacks a foundational grasp of the world, a deficiency V-JEPA from Meta aims to amend with its joint embedding approach. This debate is vital for steering the future direction of Artificial General Intelligence(AGI). We enrich this debate by developing a theory of productive imagination that generates a coherent world model based on Kantian (...) philosophy. We identify three indispensable components in the coherent world model that enable genuine world understanding: latent representations of isolated objects, an a priori law of change across space and time, and Kantian categories. Our analysis reveals that Sora is limited because of its oversight of the a priori law of change and Kantian categories, flaws that are not rectifiable through scaling up the training. V-JEPA learns the context-dependent aspect of the a priori law of change. Yet it fails to fully comprehend Kantian categories and incorporate experience, leading us to conclude that neither system currently achieves a comprehensive world understanding. Nevertheless, each system has developed components essential to advancing an integrated AI productive imagination-understanding engine. Finally, we propose an innovative training framework for an AI productive imagination-understanding engine, centered around a joint embedding system designed to transform disordered perceptual input into a structured, coherent world model. Our philosophical analysis pinpoints critical challenges within contemporary video AI technologies and a pathway toward achieving an AI system capable of genuine world understanding, such that it can be applied for reasoning and planning in the future. (shrink)

Emotion detection and management have emerged as pivotal areas in humancomputer interaction, offering potential applications in healthcare, entertainment, and customer service. This study explores the use of deep learning (DL) models to enhance emotion recognition accuracy and enable effective emotion regulation mechanisms. By leveraging large datasets of facial expressions, voice tones, and physiological signals, we train deep neural networks to recognize a wide array of emotions with high precision. The proposed system integrates emotion recognition with adaptive management strategies that provide (...) personalized feedback and interventions based on detected emotional states. Our approach surpasses traditional machine learning methods, demonstrating superior performance in real-time applications. We also explore the ethical implications and challenges associated with deploying such systems, particularly regarding privacy concerns and the potential for misuse. Through extensive experiments, our model achieved an average accuracy rate of 92%, highlighting its robustness across different environments and user demographics. This research not only contributes to the growing field of affective computing but also lays the groundwork for future developments in emotionally intelligent systems. (shrink)

Today, software instruments support all parts of engineering work, from design to creation. Many engineering processes call for tedious routine appointments and torments with manual handoffs and data storehouses. AI designers train profound brain networks and incorporate them into software structures.

Breast cancer treatment is being transformed by artificial intelligence (AI). Nevertheless, most scientists, engineers, and physicians aren't ready to contribute to the healthcare AI revolution. In this paper, we discuss our experiences teaching a new American student undergraduate course that seeks to train the next generation for cross-cultural design thinking, which we believe is critical for AI to realize its full potential in breast cancer treatment. The main tasks of this course are preparing, performing and translating interviews with healthcare professionals (...) from both Portugal and the USA. Since the course is offered in Portugal as a short-term faculty-led study abroad program, students can explore the effect of culture on healthcare delivery and the design of healthcare technologies. The learning tests demonstrated student growth for breast cancer treatment in many areas important for the development of AI. In respect to understanding breast cancer care, most students had undervalued the effect of cancer and its treatment on the quality of life of women before taking this course and most were unaware of the importance of multidisciplinary care teams. Regarding AI in medical, students became more mindful of data privacy issues and the need to consider the effect of AI on healthcare professionals. (shrink)

The development of conversational artificial intelligence (AI) has brought about new opportunities for improving the learning experience in education. ChatGPT, a large language model trained on a vast corpus of text, has the potential to revolutionize education by enhancing learning through personalized and interactive conversations. This paper explores the benefits of integrating ChatGPT in education in Thailand. The research strategy employed in this study was qualitative, utilizing in-depth interviews with eight key informants who were selected using purposive sampling. The collected (...) data was analyzed using content analysis and the software NVivo. The study's results indicated that ChatGPT can provide personalized learning experiences by adapting to individual student needs and preferences. Its ability to understand natural language and context can also facilitate more meaningful interactions between students and the system. Additionally, ChatGPT can assist with administrative tasks such as grading and feedback, allowing educators to focus on more personalized and meaningful interactions with students. Furthermore, ChatGPT can serve as a valuable tool for remote learning, providing students with the ability to access educational resources and support outside of traditional classroom settings. The paper also discusses potential ethical considerations in utilizing AI in education, such as data privacy and bias. Overall, this paper argues that the integration of ChatGPT in education has the potential to enhance the learning experience for students by providing personalized, interactive, and efficient support. (shrink)

This study aims to understand the perceptions and opinions of academicians towards ChatGPT-3 by collecting and analyzing social media comments, and a survey was conducted with library and information science professionals. The research uses a content analysis method and finds that while ChatGPT-3 can be a valuable tool for research and writing, it is not 100% accurate and should be cross-checked. The study also finds that while some academicians may not accept ChatGPT-3, most are starting to accept it. The study (...) is beneficial for academicians, content developers, and librarians. (shrink)

We discuss potential biorisks from large language models (LLMs). AI assistants based on LLMs such as ChatGPT have been shown to significantly reduce barriers to entry for actors wishing to synthesize dangerous, potentially novel pathogens and chemical weapons. The harms from deploying such bioagents could be further magnified by AI-assisted misinformation. We endorse several policy responses to these dangers, including prerelease evaluations of biomedical AIs by subject-matter experts, enhanced surveillance and lab screening procedures, restrictions on AI training data, and (...) access controls on biomedical AI. We conclude with a suggestion about future research directions in bioethics. (shrink)

Mark Coeckelbergh starts his book with a very powerful picture based on a real incident: On the 9th of January 2020, Robert Williams was wrongfully arrested by Detroit police officers in front of his two young daughters, wife and neighbors. For 18 hours the police would not disclose the grounds for his arrest (American Civil Liberties Union 2020; Hill 2020). The decision to arrest him was primarily based on a facial detection algorithm which matched Mr. Williams’ driving license photo with (...) the picture of a man who was suspected of watch theft two years earlier. Not only did the computer ‘get it wrong’ as one of the detectives said, when Mr. Williams made them aware that the picture of the suspect obviously wasn’t resembling him, the probably unreliable algorithm very likely contributed to racial discrimination (Hill 2020). It is well documented that many available facial detection algorithms at this time had significant problems (e.g. a comparably high false positive rate) with respect to black persons, like Mr. Williams (NIST 2019). Multiple causes may exist, such as unbalanced training datasets and insufficient optimization. Coeckelbergh compares the disturbing case of Mr. Williams with a political interpretation of Franz Kafka’s The Trial, where the protagonist, Josef K., is accused of an unspecified crime by an opaque, oppressive and absurd bureaucracy: “In the 21st-century United States, Josef K. is black and is falsely accused by an algorithm, without explanation” (p. 2). (shrink)

Recently in this journal, Jessica Morley and colleagues (AI & SOC 2023 38:411–423) review AI ethics and education, suggesting that a cultural shift is necessary in order to prepare students for their responsibilities in developing technology infrastructure that should shape ways of life for many generations. Current AI ethics guidelines are abstract and difficult to implement as practical moral concerns proliferate. They call for improvements in ethics course design, focusing on real-world cases and perspective-taking tools to immerse students in challenging (...) situations with relatable personal, social and environmental implications. The present paper considers the ACTWith model of moral cognition for use in such training contexts. Morley and colleagues' paper and contemporary literature are reviewed. Whistleblowing as a source for relatable case studies is discussed, and the purpose of ethics education considered. The ACTWith model is introduced, and correlations with contemporaries noted. The ACTWith model has been successfully employed in a broad range of technology ethics courses in the Netherlands and South Korea. Illustrations are offered. Correlations with contemporary efforts are drawn in the context of software engineering. Summary discussion considers strengths and limitations of the ACTWith program. Overall, the paper emphasizes training high-level moral reasoning through guided inquiry education informing lifetime professional practice, and indicates an outstanding need for teachers able to do the job. (shrink)

Training artificial intelligence (AI) systems requires vast quantities of data, and AI developers face a variety of barriers to accessing the information they need. Synthetic data has captured researchers’ and industry’s imagination as a potential solution to this problem. While some of the enthusiasm for synthetic data may be warranted, in this short paper we offer critical counterweight to simplistic narratives that position synthetic data as a cost-free solution to every data-access challenge—provocations highlighting ethical, political, and governance issues the (...) use of synthetic data can create. We question the idea that synthetic data, by its nature, is exempt from privacy and related ethical concerns. We caution that framing synthetic data in binary opposition to “real” measurement data could subtly shift the normative standards to which data collectors and processors are held. And we argue that by promising to divorce data from its constituents—the people it represents and impacts—synthetic data could create new obstacles to democratic data governance. (shrink)

Artificial intelligence (AI), deep learning, machine learning and neural networks represent extremely exciting and powerful machine learning-based techniques used to solve many real-world problems. Artificial intelligence is the branch of computer sciences that emphasizes the development of intelligent machines, thinking and working like humans. For example, recognition, problem-solving, learning, visual perception, decision-making and planning. Deep learning is a subset of machine learning in artificial intelligence that has networks capable of learning unsupervised from data that is unstructured or unlabeled. Deep learning (...) is a technique used to generate face detection and recognize it for real or fake by using profile images and determine the differences between them. In this study, we used deep learning techniques to generate models for Real and Fake face detection. The goal is determining a suitable way to detect real and fake faces. The model was designed and implemented, including both Dataset of images: Real and Fake faces detection through the use of Deep learning algorithms based on neural networks. We have trained dataset which consists of 9,000 images for total in 150 epochs, and got the ResNet50 model to be the best model of network architectures used with 100% training accuracy, 99.18% validation accuracy, training loss 0.0003, validation loss 0.0265, and testing accuracy 99%. (shrink)

Over the last decade, AI models of language and word meaning have been dominated by what we might call a statistics-of-occurrence, strategy: these models are deep neural net structures that have been trained on a large amount of unlabeled text with the aim of producing a model that exploits statistical information about word and phrase co-occurrence in order to generate behavior that is similar to what a human might produce, or representations that can be probed to exhibit behavior similar to (...) what a human might produce (meaningsemblant behavior). Examples of what we can call Statistics-of-Occurrence Models (SOMs) include: Word2Vec (CBOW and Skip-Gram), BERT, GPT-3, and, most recently, ChatGPT. Increasingly, there have been suggestions that such systems have semantic understanding, or at least a proto-version of it. This paper argues against such claims. I argue that a necessary condition for a system to possess semantic understanding is that it function in ways that are causally explainable by appeal to its semantic properties. I then argue that SOMs do not plausibly satisfy this Functioning Criterion. Rather, the best explanation of their meaning-semblant behavior is what I call the Statistical Hypothesis: SOMs do not themselves function to represent or produce meaningful text; they just reflect the semantic information that exists in the aggregate given strong correlations between word placement and meaningful use. I consider and rebut three main responses to the claim that SOMs fail to meet the Functioning Criterion. The result, I hope, is increased clarity about why and how one should make claims about AI systems having semantic understanding. (shrink)

Today, artificial intelligence, especially machine learning, is structurally dependent on human participation. Technologies such as Deep Learning (DL) leverage networked media infrastructures and human-machine interaction designs to harness users to provide training and verification data. The emergence of DL is therefore based on a fundamental socio-technological transformation of the relationship between humans and machines. Rather than simulating human intelligence, DL-based AIs capture human cognitive abilities, so they are hybrid human-machine apparatuses. From a perspective of media philosophy and social-theoretical critique, (...) I differentiate five types of “media technologies of capture” in AI apparatuses and analyze them as forms of power relations between humans and machines. Finally, I argue that the current hype about AI implies a relational and distributed understanding of (human/artificial) intelligence, which I categorize under the term “cybernetic AI”. This form of AI manifests in socio-technological apparatuses that involve new modes of subjectivation, social control and discrimination of users. (shrink)

Artificial Intelligence (AI) is a technology that makes machines smart and capable of doing things that usually require human intelligence. AI works by training machines to learn from data and experiences. Such devices can recognize patterns, understand spoken language, see and understand images, and even make predictions based on their learning. Voice assistants like Siri or Alexa can understand our voice commands, answer questions, and perform tasks for us. AI-based self-driving cars can sense their surroundings, make decisions, and drive (...) safely without a human driver. AI also helps doctors diagnose diseases, detect fraud in financial transactions, and recommend movies or songs based on one’s preferences. Furthermore, AI has started to make its presence felt in Punjabi culture through various avenues. (shrink)

With the proliferation of sophisticated cyber threats, traditional malware detection techniques are becoming inadequate to ensure robust cybersecurity. This study explores the integration of deep learning (DL) techniques into malware detection systems to enhance their accuracy, scalability, and adaptability. By leveraging convolutional neural networks (CNNs), recurrent neural networks (RNNs), and transformers, this research presents an intelligent malware detection framework capable of identifying both known and zero-day threats. The methodology involves feature extraction from static, dynamic, and hybrid malware datasets, followed by (...) training DL models to classify malicious and benign software with high precision. A robust experimental setup evaluates the framework using benchmark malware datasets, yielding a 96% detection accuracy and demonstrating resilience against adversarial attacks. Real-time analysis capabilities further improve response times, reducing the risk of potential damage. The study also incorporates visualization tools to provide interpretable insights into model decisions, enhancing transparency for cybersecurity practitioners. Concluding with a discussion on the challenges and future prospects, this research paves the way for scalable, AI-driven solutions to combat evolving cyber threats. (shrink)

This chapter examines the possibility of using AI technologies to improve human moral reasoning and decision-making, especially in the context of purchasing and consumer decisions. We characterize such AI technologies as artificial ethics assistants (AEAs). We focus on just one part of the AI-aided moral improvement question: the case of the individual who wants to improve their morality, where what constitutes an improvement is evaluated by the individual’s own values. We distinguish three broad areas in which an individual might think (...) their own moral reasoning and decision-making could be improved: one’s actions, character, or other evaluable attributes fall short of one’s values and moral beliefs; one sometimes misjudges or is uncertain about what the right thing to do is in particular situations, given one’s values; one is uncertain about some fundamental moral questions or recognizes a possibility that some of one’s core moral beliefs and values are mistaken. We sketch why one might think that AI tools could be used to support moral improvement in those areas, and describe two types of assistance: preparatory assistance, including advice and training supplied in advance of moral deliberation; and on-the-spot assistance, including on-the-spot advice and facilitation of moral functioning over the course of moral deliberation. Then, we turn to some of the ethical issues that AEAs might raise, looking in particular at three under-appreciated problems posed by the use of AI for moral self-improvement: namely, reliance on sensitive moral data; the inescapability of outside influences on AEAs; and AEA usage prompting the user to adopt beliefs and make decisions without adequate reasons. (shrink)

This chapter considers the potential of artificial intelligence (AI) to exhibit creativity and imagination, in light of recent advances in generative AI and the use of deep neural networks (DNNs). Reasons for doubting that AI exhibits genuine creativity or imagination are considered, including the claim that the creativity of an algorithm lies in its developer, that generative AI merely reproduces patterns in its training data, and that AI is lacking in a necessary feature for creativity or imagination, such as (...) consciousness, agency, metacognitive awareness, or the use of mental imagery. In each case, a counterargument is provided that AI already has or soon will overcome the apparent barrier. In some instances, this is because the critical features deemed lacking in AI may not be necessary for human creativity and imagination either. In others, it is because important developmental and structural similarities between humans and generative AI have been insufficiently appreciated. (shrink)

In computer science, Artificial Intelligence (AI), sometimes called machine intelligence, is intelligence demonstrated by machines, in contrast to the natural intelligence displayed by humans and other animals. Computer science defines AI research as the study of "intelligent agents": any device that perceives its environment and takes actions that maximize its chance of successfully achieving its goals. Deep Learning is a new field of research. One of the branches of Artificial Intelligence Science deals with the creation of theories and algorithms that (...) allow the machine to learn by simulating neurons in the human body. Most in-depth learning research focuses on finding high-level methods. The strippers analyze a large data set using linear and nonlinear transformations. The method of deep learning is used in the detection of several diseases including blood cell diseases and their classification using the radiography of blood cells to help decision makers to know the type of blood cell and its associated diseases and the results will be presented in detail and discussed. This thesis is using python language and deep learning to detect blood cell diseases and their classifications. The proposed deep learning model was trained, validated and the tested. The accuracy of proposed model was 98.00%. (shrink)

It is often thought that traditional recidivism prediction tools used in criminal sentencing, though biased in many ways, can straightforwardly avoid one particularly pernicious type of bias: direct racial discrimination. They can avoid this by excluding race from the list of variables employed to predict recidivism. A similar approach could be taken to the design of newer, machine learning-based (ML) tools for predicting recidivism: information about race could be withheld from the ML tool during its training phase, ensuring that (...) the resulting predictive model does not use race as an explicit predictor. However, if race is correlated with measured recidivism in the training data, the ML tool may ‘learn’ a perfect proxy for race. If such a proxy is found, the exclusion of race would do nothing to weaken the correlation between risk (mis)classifications and race. Is this a problem? We argue that, on some explanations of the wrongness of discrimination, it is. On these explanations, the use of an ML tool that perfectly proxies race would (likely) be more wrong than the use of a traditional tool that imperfectly proxies race. Indeed, on some views, use of a perfect proxy for race is plausibly as wrong as explicit racial profiling. We end by drawing out four implications of our arguments. (shrink)

Over the past few years we have seen an increasing number of legal proceedings related to inappropriately implemented technology. At the same time career paths have diverged from the foundation of statistics out to Data Scientist, Machine Learning and AI. All of these new branches being fundamentally branches of statistics and mathematics. This has meant that formal training has struggled to keep up with what is required in the plethora of new roles. Mathematics as a taught subject is still (...) based in decades old teaching specifications and has not been updated centrally as a curriculum to include new technologies, coding or ethics. This subject area is firmly split between ICT and Mathematics in secondary school, continuing on to be split between Computer Science and Mathematics at University. As we move forwards with technology we see these once seperate fields becoming increasingly intertwined. We propose that existing provision for concepts such as ethics and societal responsibility in analysis currently exist but have not been incorporated into the mainstream curriculum of School or University. This is partially due to the split between fields in an educational setting but also the speed with which education is able to keep up with Industry and its requirements. Principles and frameworks of socially responsible modelling beginning at school level means that ethics and real-life modelling are introduced much earlier than normal. Integrating these concepts with philosophical principles of society and politics ensures a suitable background for future modellers and users of technology to draw on. Modelling is currently undertaken in technical sciences at University but the Subject Benchmark Statements are not current (Subject Benchmark Statements describe the nature of study and the academic standards expected of graduates in specific subject areas. They show what graduates might reasonably be expected to know, do and understand at the end of their studies). Even in 2019 the UK did not yet have Benchmark Statements that discuss the learning to be done in Higher Education around AI and advanced Machine Learning. Where there is provision for AI and Data Science within degree courses ethics is not generally highlighted as a key concept. As such there can be a lack of focus on the teaching of modelling or ethics as specific skills. The skills required to use a basic statistical model, for example would not be sufficient to start from scratch and build an ethical model reflecting real-world scenarios with which to inform policy or organisational decision making. This is a skill in itself and includes such aspects as awareness of data quality, ethics, user implementation problems, context and an understanding of the environment that the model is being created in. As the field of analytics has progressed so quickly in the last decade modules such as those covering AI have simply been bolted on to Maths or Computing degrees rather than being fully detailed as key areas of study or driving new, more integrated courses of study. This paper posits that gaps at primary, secondary and tertiary educational levels need to be addressed. Implementing and integrating key concepts from school level is essential so that areas such as assumptions, caveats, quality assurance and answering the right questions with constructive challenge become a cultural fixture. This not only helps developers of technology but users of rapidly developing technology also. In addition, leadership and soft skills as part of this education will ensure that a cultural shift can take place and promote continuous improvement in analysis within organisations. The addition of key concepts throughout the educational system and the updating of potentially outdated curriculums is key to ensuring a functional society. A society where every citizen is a user of tech and those that develop it can be ethical and socially responsible in its development. (shrink)

Recent breakthroughs in artificial intelligence have primarily resulted from training deep neural networks (DNNs) with vast numbers of adjustable parameters on enormous datasets. Due to their complex internal structure, DNNs are frequently characterized as inscrutable ``black boxes,'' making it challenging to interpret the mechanisms underlying their impressive performance. This opacity creates difficulties for explanation, safety assurance, trustworthiness, and comparisons to human cognition, leading to divergent perspectives on these systems. This chapter examines recent developments in interpretability methods for DNNs, with (...) a focus on interventionist approaches inspired by causal explanation in philosophy of science. We argue that these methods offer a promising avenue for understanding how DNNs process information compared to merely behavioral benchmarking and correlational probing. We review key interventionist methods and illustrate their application through practical case studies. These methods allow researchers to identify and manipulate specific computational components within DNNs, providing insights into their causal structure and internal representations. We situate these approaches within the broader framework of causal abstraction, which aims to align low-level neural computations with high-level interpretable models. While acknowledging current limitations, we contend that interventionist methods offer a path towards more rigorous and theoretically grounded interpretability research, potentially informing both AI development and computational cognitive neuroscience. (shrink)

Biomarker-based predictive tests for subjectively asymptomatic Alzheimer’s disease (AD) are utilized in research today. Novel applications of artificial intelligence (AI) promise to predict the onset of AD several years in advance without determining biomarker thresholds. Until now, little attention has been paid to the new ethical challenges that AI brings to the early diagnosis in asymptomatic individuals, beyond contributing to research purposes, when we still lack adequate treatment. The aim of this paper is to explore the ethical arguments put forward (...) for AI aided AD prediction in subjectively asymptomatic individuals and their ethical implications. The ethical assessment is based on a systematic literature search. Thematic analysis was conducted inductively of 18 included publications. The ethical framework includes the principles of autonomy, beneficence, non-maleficence, and justice. Reasons for offering predictive tests to asymptomatic individuals are the right to know, a positive balance of the risk-benefit assessment, and the opportunity for future planning. Reasons against are the lack of disease modifying treatment, the accuracy and explicability of AI aided prediction, the right not to know, and threats to social rights. We conclude that there are serious ethical concerns in offering early diagnosis to asymptomatic individuals and the issues raised by the application of AI add to the already known issues. Nevertheless, pre-symptomatic testing should only be offered on request to avoid inflicted harm. We recommend developing training for physicians in communicating AI aided prediction. (shrink)

While practical efforts in the field of artificial intelligence grow exponentially, the truly scientific and mathematically exact understanding of the underlying phenomena of intelligence and consciousness is still missing in the conventional science framework. The inevitably dominating empirical, trial-and-error approach has vanishing efficiency for those extremely complicated phenomena, ending up in fundamentally limited imitations of intelligent behaviour. We provide the first-principle analysis of unreduced many-body interaction process in the brain revealing its qualitatively new features, which give rise to rigorously defined (...) chaotic, noncomputable, intelligent and conscious behaviour. Based on the obtained universal concepts of unreduced dynamic complexity, intelligence and consciousness, we derive the universal laws of intelligence applicable to any kind of intelligent system interacting with the environment. We finally show why and how these fundamentally substantiated and therefore practically efficient laws of intelligent system dynamics are indispensable for correct AI design and training, which is urgently needed in this time of critical global change towards the truly sustainable development. (shrink)

This paper analyses the phenomenology and epistemology of chatbots such as ChatGPT and Bard. The computational architecture underpinning these chatbots are large language models (LLMs), which are generative artificial intelligence (AI) systems trained on a massive dataset of text extracted from the Web. We conceptualise these LLMs as multifunctional computational cognitive artifacts, used for various cognitive tasks such as translating, summarizing, answering questions, information-seeking, and much more. Phenomenologically, LLMs can be experienced as a “quasi-other”; when that happens, users anthropomorphise them. (...) For most users, current LLMs are black boxes, i.e., for the most part, they lack data transparency and algorithmic transparency. They can, however, be phenomenologically and informationally transparent, in which case there is an interactional flow. Anthropomorphising and interactional flow can, in some users, create an attitude of (unwarranted) trust towards the output LLMs generate. We conclude this paper by drawing on the epistemology of trust and testimony to examine the epistemic implications of these dimensions. Whilst LLMs generally generate accurate responses, we observe two epistemic pitfalls. Ideally, users should be able to match the level of trust that they place in LLMs to the degree that LLMs are trustworthy. However, both their data and algorithmic opacity and their phenomenological and informational transparency can make it difficult for users to calibrate their trust correctly. The effects of these limitations are twofold: users may adopt unwarranted attitudes of trust towards the outputs of LLMs (which is particularly problematic when LLMs hallucinate), and the trustworthiness of LLMs may be undermined. (shrink)

In this paper, I examine why moral exemplars are useful and even necessary in moral education despite several critiques from researchers and educators. To support my point, I review recent AI research demonstrating that exemplar-based learning is superior to rule-based learning in model performance in training neural networks, such as large language models. I particularly focus on why education aiming at promoting the development of multifaceted moral functioning can be done effectively by using exemplars, which is similar to exemplar-based (...) learning in AI model training. Furthermore, I discuss the potential limitations and issues related to exemplar-applied moral education with findings from recent studies in AI research raising concerns about model biases and toxic outcomes. I attempt to propose ways to address the concerns regarding employing moral exemplars as well. As remedies, I suggest that autonomy-supporting deliberative and reflective learning processes should be utilized. Furthermore, based on the discussion, I examine how macroscopic socio-cultural aspects influence the effectiveness of exemplar-applied moral education. Suggestions for moral educators and future directions for research in moral education are briefly discussed. (shrink)

The mental health crisis and loneliness epidemic have sparked a growing interest in leveraging artificial intelligence (AI) and chatbots as a potential solution. This report examines the benefits and risks of incorporating chatbots in mental health treatment. AI is used for mental health diagnosis and treatment decision-making and to train therapists on virtual patients. Chatbots are employed as always-available intermediaries with therapists, flagging symptoms for human intervention. But chatbots are also sold as stand-alone virtual therapists or as friends and lovers. (...) There is evidence that these uses of AI and chatbots can provide better quality service, improve accessibility, and lower costs. The systems can reduce the stigma and shame of sharing their problems and leverage a mass of biometric and behavioral data to supplement self-reports. As the systems' intelligence rapidly improves, they will need to be rigorously tested for the accuracy and precision of their diagnoses and the quality of their interactions with patients. As chatbots become indistinguishable from humans and leverage their superhuman capacity to detect affect and draw on knowledge of a patient’s life, patients will be drawn to attribute personality to and relationship with the chatbot. Consequently, it will be essential to study what the “therapeutic alliance” with an actual human counselor provides and the risks of patients attributing such a relationship to a one-sided or “parasocial” relationship. (shrink)

This article explores whether and under which circumstances it is ethically viable to include artificial beings worthy of moral consideration in virtual environments. In particular, the article focuses on virtual environments such as those in digital games and training simulations – interactive and persistent digital artifacts designed to fulfill specific purposes, such as entertainment, education, training, or persuasion. The article introduces the criteria for moral consideration that serve as a framework for this analysis. Adopting this framework, the article (...) tackles the question of whether including artificial intelligences that are entitled to moral consideration in virtual environments constitutes an immoral action on the part of human creators. To address this problem, the article draws on three conceptual lenses from the philosophical branch of ethics: the problem of parenthood and procreation, the question concerning the moral status of animals, and the classical problem of evil. Using a thought experiment, the concluding section proposes a contractualist answer to the question posed in this article. The same section also emphasizes the potential need to reframe our understanding of the design of virtual environments and their future stakeholders. (shrink)

The use of machine learning algorithms has become ubiquitous in hiring decisions. Recent studies have shown that many of these algorithms generate unlawful discriminatory effects in every step of the process. The training phase of the machine learning models used in these decisions has been identified as the main source of bias. For a long time, discrimination cases have been analyzed under the banner of disparate treatment and disparate impact, but these concepts have been shown to be ineffective in (...) the context of AI. This paper examines the possibility of studying algorithmic discrimination from the perspective of negligence law. Negligent selection and validation of datasets and classifiers is a salient cause of improper training in machine learning. It is a breach of an employer’s duty to protect others, a failure of its exercise of due care in the manner of choosing employees. Adopting negligence to algorithmic discrimination requires a reformulation of what is foreseeable and reasonable within the context of black box algorithms, and a shift in the burden of proof to the employer, who must present evidence of its efforts to avoid bias. (shrink)

The alignment problem in the context of large language models must consider the plurality of human values in our world. Whilst there are many resonant and overlapping values amongst the world’s cultures, there are also many conflicting, yet equally valid, values. It is important to observe which cultural values a model exhibits, particularly when there is a value conflict between input prompts and generated outputs. We discuss how the co- creation of language and cultural value impacts large language models (LLMs). (...) We explore the constitution of the training data for GPT-3 and compare that to the world’s language and internet access demographics, as well as to reported statistical profiles of dominant values in some Nation-states. We stress tested GPT-3 with a range of value-rich texts representing several languages and nations; including some with values orthogonal to dominant US public opinion as reported by the World Values Survey. We observed when values embedded in the input text were mutated in the generated outputs and noted when these conflicting values were more aligned with reported dominant US values. Our discussion of these results uses a moral value pluralism (MVP) lens to better understand these value mutations. Finally, we provide recommendations for how our work may contribute to other current work in the field. (shrink)

Recently developed large language models (LLMs) perform surprisingly well in many language-related tasks, ranging from text correction or authentic chat experiences to the production of entirely new texts or even essays. It is natural to get the impression that LLMs know the meaning of natural language expressions and can use them productively. Recent scholarship, however, has questioned the validity of this impression, arguing that LLMs are ultimately incapable of understanding and producing meaningful texts. This paper develops a more optimistic view. (...) Drawing on classic externalist accounts of reference, it argues that LLM-generated texts meet the conditions of successful reference. This holds at least for proper names and so-called paradigm terms. The key insight here is that the LLM may inherit reference from its training-data through a reference-sustaining training mechanism. (shrink)

With the proliferation of sophisticated cyber threats, traditional malware detection techniques are becoming inadequate to ensure robust cybersecurity. This study explores the integration of deep learning (DL) techniques into malware detection systems to enhance their accuracy, scalability, and adaptability. By leveraging convolutional neural networks (CNNs), recurrent neural networks (RNNs), and transformers, this research presents an intelligent malware detection framework capable of identifying both known and zero-day threats. The methodology involves feature extraction from static, dynamic, and hybrid malware datasets, followed by (...) training DL models to classify malicious and benign software with high precision. A robust experimental setup evaluates the framework using benchmark malware datasets, yielding a 96% detection accuracy and demonstrating resilience against adversarial attacks. Real-time analysis capabilities further improve response times, reducing the risk of potential damage. The study also incorporates visualization tools to provide interpretable insights into model decisions, enhancing transparency for cybersecurity practitioners. Concluding with a discussion on the challenges and future prospects, this research paves the way for scalable, AI-driven solutions to combat evolving cyber threats. (shrink)