Search

In the classical Turing test, participants are challenged to tell whether they are interacting with another human being or with a machine. The way the interaction takes place is not direct, but a distant conversation through computer screen messages. Basic forms of interaction are face-to-face and embodied, context-dependent and based on the detection of reciprocal sensorimotor contingencies. Our idea is that interaction detection requires the integration of proprioceptive and interoceptive patterns with sensorimotor patterns, within quite short time lapses, so that they appear as mutually contingent, as reciprocal. In other words, the experience of interaction takes place when sensorimotor patterns are contingent upon one’s own movements, and vice versa. I react to your movement, you react to mine. When I notice both components, I come to experience an interaction. Therefore, we designed a “minimal” Turing test to investigate how much information is required to detect these reciprocal sensorimotor contingencies. Using a new version of the perceptual crossing paradigm, we tested whether participants resorted to interaction detection to tell apart human from machine agents in repeated encounters with these agents. In two studies, we presented participants with movements of a human agent, either online or offline, and movements of a computerized oscillatory agent in three different blocks. In each block, either auditory or audiovisual feedback was provided along each trial. Analysis of participants’ explicit responses and of the implicit information subsumed in the dynamics of their series will reveal evidence that participants use the reciprocal sensorimotor contingencies within short time windows. For a machine to pass this minimal Turing test, it should be able to generate this sort of reciprocal contingencies. [ABSTRACT FROM AUTHOR]

We start from the fundamental premise that any physical interaction can be interpreted as a game. To demonstrate this, we draw upon the free energy principle and the theory of quantum reference frames. In this way, we place the game-theoretic Nash Equilibrium in a new light in so far as the incompleteness and undecidability of the concept, as well as the nature of strategies in general, can be seen as the consequences of certain no-go theorems. We show that games of the generic imitation type follow a circularity of idealization that includes the good regulator theorem, generalized synchrony, and undecidability of the Turing test. We discuss Bayesian games in the light of Bell non-locality and establish the basics of quantum games, which we relate to local operations and classical communication protocols. In this light, we also review the rationality of gaming strategies from the players' point of view. [ABSTRACT FROM AUTHOR]

The recent period has witnessed automated crawlers designed to automatically crack passwords, which greatly risks various aspects of our lives. To prevent passwords from being cracked, image verification codes have been implemented to accomplish the human–machine verification. It is important to note, however, that the most widely‐used image verification codes, especially the visual reasoning Completely Automated Public Turing tests to tell Computers and Humans Apart (CAPTCHAs), are still susceptible to attacks by artificial intelligence. Taking the visual reasoning CAPTCHAs representing the image verification codes, this study introduces an enhanced approach for generating image verification codes and proposes an improved Convolutional Neural Network (CNN)‐based recognition system. After we add a fully connected layer and briefly solve the edge of stability issue, the accuracy of the improved CNN model can smoothly approach 98.40% within 50 epochs on the image verification codes with four digits using a large initial learning rate of 0.01. Compared with the baseline model, it is approximately 37.82% better in accuracy without obvious curve oscillation. The improved CNN model can also smoothly reach the accuracy of 99.00% within 7500 epochs on the image verification codes with six characters, including digits, upper‐case alphabets, lower‐case alphabets, and symbols. A detailed comparison between our proposed approach and the baseline one is presented. The relationship between the time consumption and the length of the seeds is compared theoretically. Subsequently, we figure out the threat assignments on the visual reasoning CAPTCHAs with different lengths based on four machine learning models. Based on the threat assignments, the Kaplan‐Meier (KM) curves are computed. [ABSTRACT FROM AUTHOR]

This paper makes the case for a sharper terminology regarding AIs cognitive abilities. In arguing that thinking requires more than content production, I offer a definition of meaning drawing on a clear distinction between living and machine intelligence. A pivotal argument is the re-use of the Turing Test (TT) for understanding which theories of meaning and consciousness are no longer plausible—because they have been reproduced by software without thereby gaining conscious experience. In following the few theories that have not (yet) failed this reversed Turing Test (RTT), the focus turns towards rethinking the human condition in times of AI along the lines of three questions: What if a machine developed consciousness? What if AI proceeded without developing a consciousness? What, if machinic and human intelligence merged? These three questions in the end lead to examining three related possible futures of humanism as now determined by the relation between Human Intelligence and AI. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) research aims to model the way human thinking works, in an attempt to reproduce abilities such as learning, decision-making and problem-solving. The concept of artificial intelligence is not a new one, research on it began in the 1950s when Alan Turing formulated the Turing test, a test to measure the intelligence of a machine. Major milestones in AI research include the creation of the first artificial neural networks in the 1950s and the first game programs in the 1960s. In the 1970s, the algorithms and techniques that still form the basis of AI today, such as machine learning and symbolic AI, began to emerge. The real breakthrough came with the integration of AI into everyday use in the 1990s and early 2000s. With the rise of the internet and the digitalization of data, AI has spread to a wide range of areas, including internet search engines, recommendation systems, online shopping algorithms and virtual assistants in smartphones. In the 2010s, AI continued to make significant progress in areas such as autonomous vehicles, medicine and robotics. AIbased language models, such as the GPT-3.5 used by ChatGPT, later upgraded to GPT-4 base models, published in November 2022, are increasingly becoming part of our daily lives and across industries are being used in a growing number of fields to facilitate intelligent communication and humanmachine interaction. In this paper, we are focusing on AI-based support for teachers. We will explore areas where AI can be effectively used to increase the efficiency of teachers' work, enumerating some AIbased solutions that can greatly contribute to relieving teachers of their daily routine tasks. [ABSTRACT FROM AUTHOR]

The objectives of this study were to create a mutual conversion system between contrast-enhanced computed tomography (CECT) and non-CECT images using a cycle generative adversarial network (cycleGAN) for the internal jugular region. Image patches were cropped from CT images in 25 patients who underwent both CECT and non-CECT imaging. Using a cycleGAN, synthetic CECT and non-CECT images were generated from original non-CECT and CECT images, respectively. The peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) were calculated. Visual Turing tests were used to determine whether oral and maxillofacial radiologists could tell the difference between synthetic versus original images, and receiver operating characteristic (ROC) analyses were used to assess the radiologists' performances in discriminating lymph nodes from blood vessels. The PSNR of non-CECT images was higher than that of CECT images, while the SSIM was higher in CECT images. The Visual Turing test showed a higher perceptual quality in CECT images. The area under the ROC curve showed almost perfect performances in synthetic as well as original CECT images. In conclusion, synthetic CECT images created by cycleGAN appeared to have the potential to provide effective information in patients who could not receive contrast enhancement. [ABSTRACT FROM AUTHOR]

Visual question answering (VQA) research has garnered increasing attention in recent years. It is considered a visual Turing test because it requires a computer to respond to textual questions based on an image. Expertise in computer vision, natural language processing, knowledge understanding, and reasoning is required to solve the problem of VQA. Most techniques employed for VQA consist of models that are developed to learn the combination of image and question features along with the expected answer. The techniques chosen for image and question feature extraction and combining the features change with each model. This method of teaching a model of the question–answer pattern is ineffective for queries that involve counting and reasoning. This approach also requires considerable resources and large datasets for the training. The general VQA datasets feature a restricted number of items as responses to counting questions (< 1 0), and the distribution of the answers is not uniform. To investigate these issues in VQA, we created synthetic datasets that could be modified to adjust the number of objects in the image and the amount of occlusion. Specifically, a zero-shot learning VQA system was devised for counting-related questions that provide answers by analyzing the output of an object detector and the query keywords. Using synthetic datasets, our model generated 100% correct results. Testing on the benchmark datasets task directed image understanding challenge (TDIUC) and TallyQA-simple indicated that the proposed model matched the performance of the learning-based baseline models. This methodology can be used efficiently for counting VQA questions confined to certain domains when the number of items to be counted is significant. [ABSTRACT FROM AUTHOR]

Generative art has drawn increased attention in recent AI applications. Traditional approaches of robot calligraphy have faced challenges in achieving style consistency, line smoothness and high-quality structural uniformity. To address the limitation of existing methods, we propose a dual generator framework based on deep adversarial networks for robotic calligraphy reproduction. The proposed model utilizes a encoder-decoder module as one generator for style learning and a robot arm as the other generator for motion learning to optimize the networks and obtain the best robot calligraphy works. Based on the enhanced datasets, multiple evaluation metrics including coverage rate, structural similarity index measure, intersection over union and Turing test are employed to perform the experimental validation. The evaluations demonstrate that the proposed method is highly effective and applicable in robot calligraphy and achieves state-of-the-art results with the average structural similarity index measure 75.91% , coverage rate 70.25%, and intersection over union 80.68%, which provides a paradigm for evaluation in the field of art. [ABSTRACT FROM AUTHOR]

Abstract This retrospective, multi-centered study aimed to improve high-quality radiation treatment (RT) planning workflows by training and testing a Convolutional Neural Network (CNN) to perform auto segmentations of organs at risk (OAR) for prostate cancer (PCa) patients, specifically the bladder and rectum. The objective of this project was to develop a clinically applicable and robust artificial intelligence (AI) system to assist radiation oncologists in OAR segmentation. The CNN was trained using manual contours in CT-datasets from diagnostic 68Ga-PSMA-PET/CTs by a student, then validated (n = 30, PET/CTs) and tested (n = 16, planning CTs). Further segmentations were generated by a commercial artificial intelligence (cAI) software. The ground truth were manual contours from expert radiation oncologists. The performance was evaluated using the Dice-Sørensen Coefficient (DSC), visual analysis and a Turing test. The CNN yielded excellent results in both cohorts and OARs with a DSCmedian > 0.87, the cAI resulted in a DSC > 0.78. In the visual assessment, 67% (bladder) and 75% (rectum) of the segmentations were rated as acceptable for treatment planning. With a misclassification rate of 45.5% (bladder) and 51.1% (rectum), the CNN passed the Turing test. The metrics, visual assessment and the Turing test confirmed the clinical applicability and therefore the support in clinical routine.

Bayes factor hypothesis testing provides a powerful framework for assessing the evidence in favor of competing hypotheses. To obtain Bayes factors, statisticians often require advanced, non-standard tools, making it important to confirm that the methodology is computationally sound. This paper seeks to validate Bayes factor calculations by applying two theorems attributed to Alan Turing and Jack Good. The procedure entails simulating data sets under two hypotheses, calculating Bayes factors, and assessing whether their expected values align with theoretical expectations. We illustrate this method with an ANOVA example and a network psychometrics application, demonstrating its efficacy in detecting calculation errors and confirming the computational correctness of the Bayes factor results. This structured validation approach aims to provide researchers with a tool to enhance the credibility of Bayes factor hypothesis testing, fostering more robust and trustworthy scientific inferences. [ABSTRACT FROM AUTHOR]

Online reviews serve as a guide for consumer choice. With advancements in large language models (LLMs) and generative AI, the fast and inexpensive creation of human-like text may threaten the feedback function of online reviews if neither readers nor platforms can differentiate between human-written and AI-generated content. In two experiments, we found that humans cannot recognize AI-written reviews. Even with monetary incentives for accuracy, both Type I and Type II errors were common: human reviews were often mistaken for AI-generated reviews, and even more frequently, AI-generated reviews were mistaken for human reviews. This held true across various ratings, emotional tones, review lengths, and participants' genders, education levels, and AI expertise. Younger participants were somewhat better at distinguishing between human and AI reviews. An additional study revealed that current AI detectors were also fooled by AI-generated reviews. We discuss the implications of our findings on trust erosion, manipulation, regulation, consumer behavior, AI detection, market structure, innovation, and review platforms. [ABSTRACT FROM AUTHOR]

Taking the deictic judgment that is the modernist gesture of declaring something to be art as a starting point, this essay suggests an analogous deixis as a necessary condition for literature. This deixis also can serve as the basis for discussing the expectations of computer‐generated texts. Against the idea that computers or AI systems need only produce sufficiently good output in order to be considered authors, the essay proposes an approach that takes the social recognition of the deictic act within a community of judgment as a precondition for authorship. As an alternative to the Turing test, which is based on the paradigm of deception (people are tricked into considering computer‐written text to be written by humans), the essay favors a version of Susan Leigh Star's "Durkheim test," which is based on the paradigm of co‐sociality (people directly recognize computers as social actors). Only if the gesture of a machine declaring something to be art is recognized as a deictic judgment in the full sense can one plausibly speak of computer authorship. [ABSTRACT FROM AUTHOR]

Since the 1950s, when the Turing Test was introduced, there has been notable progress in machine language intelligence. Language modeling, crucial for AI development, has evolved from statistical to neural models over the last two decades. Recently, transformer-based Pre-trained Language Models (PLM) have excelled in Natural Language Processing (NLP) tasks by leveraging large-scale training corpora. Increasing the scale of these models enhances performance significantly, introducing abilities like context learning that smaller models lack. The advancement in Large Language Models, exemplified by the development of ChatGPT, has made significant impacts both academically and industrially, capturing widespread societal interest. This survey provides an overview of the development and prospects from Large Language Models (LLM) to Large Multimodal Models (LMM). It first discusses the contributions and technological advancements of LLMs in the field of natural language processing, especially in text generation and language understanding. Then, it turns to the discussion of LMMs, which integrates various data modalities such as text, images, and sound, demonstrating advanced capabilities in understanding and generating cross-modal content, paving new pathways for the adaptability and flexibility of AI systems. Finally, the survey highlights the prospects of LMMs in terms of technological development and application potential, while also pointing out challenges in data integration, cross-modal understanding accuracy, providing a comprehensive perspective on the latest developments in this field. [ABSTRACT FROM AUTHOR]

With the rapid development of machine intelligence, an increasing number of websites and servers have been amicably visited or sometimes attached by intelligent machines intensively. Therefore, how to empower a host machine to intelligently distinguish intelligent machines from humans is a challenging work. In this paper, the Mirror Turing Test (MTT) is conceived and implemented. Unlike the standard Turing Test, the tester in the MTT is replaced by a machine instead of a human. Current advancements on deep learning enable machines to recognize subtle differences between genuine and counterfeit works. Sometimes, the ability of machines is even superior to that of humans. Will machines transcend humans in an irreversible trend? Not completely right. The detection of soul in an artwork remains far beyond the capacity of machines. The two sets of MTT based on poetry generated by a machine and a novel imitated by a human were conducted in this paper and neither of them passed the MTT. Poetry is one of the art forms in which authors reveal their souls. Thus, we chose poetry in the MTT experiments on the basis of our soul computing model, thus clearly discriminating machine from human. [ABSTRACT FROM AUTHOR]

Computed tomography (CT) and magnetic resonance imaging (MRI) are widely utilized modalities for primary clinical imaging, providing crucial anatomical and pathological information for diagnosis. CT measures X‐ray attenuation, while MRI captures hydrogen atom density in tissues. Despite their distinct imaging physics principles, the signals obtained from both modalities when imaging the same subject can be represented by modality‐specific parameters and common latent variables related to anatomy and pathology. This paper proposes an adversarial learning approach using deep convolutional neural networks to disentangle these factors. This disentanglement allows us to simulate one modality from the other. Experimental results demonstrate our ability to generate synthetic CT images from MRI inputs using the Gold‐atlas dataset, which consists of paired CT‐MRI volumes. Patch‐based learning techniques and a visual Turing test are employed to model discriminator losses. Our approach achieves a mean absolute error of μ±σ$$ \left(\mu \pm \sigma \right) $$ 36.81 ±$$ \pm $$ 4.46 HU, peak signal to noise ratio of 26.12 ±$$ \pm $$ 0.31 dB, and structural similarity measure of 0.9 ±$$ \pm $$ 0.02. Notably, the synthetic CT images accurately represent bones, gaseous cavities, and soft tissue textures, which can be challenging to visualize in MRI. The proposed model operates at an inference compute cost of 430.68 GFlops/voxel. This method can minimize radiation exposure by reducing the need for pre‐operative CT scans, providing an MR‐only alternative in clinical settings. [ABSTRACT FROM AUTHOR]

As a Turing test in multimedia, visual question answering (VQA) aims to answer the textual question with a given image. Recently, the "dynamic" property of neural networks has been explored as one of the most promising ways of improving the adaptability, interpretability, and capacity of the neural network models. Unfortunately, despite the prevalence of dynamic convolutional neural networks, it is relatively less touched and very nontrivial to exploit dynamics in the transformers of the VQA tasks through all the stages in an end-to-end manner. Typically, due to the large computation cost of transformers, researchers are inclined to only apply transformers on the extracted high-level visual features for downstream vision and language tasks. To this end, we introduce a question-guided dynamic layer to the transformer as it can effectively increase the model capacity and require fewer transformer layers for the VQA task. In particular, we name the dynamics in the Transformer as Conditional Multi-Head Self-Attention block (cMHSA). Furthermore, our questionguided cMHSA is compatible with conditional ResNeXt block (cResNeXt). Thus a novel model mixture of conditional gating blocks (McG) is proposed for VQA, which keeps the best of the Transformer, convolutional neural network (CNN), and dynamic networks. The pure conditional gating CNN model and the conditional gating Transformer model can be viewed as special examples of McG. We quantitatively and qualitatively evaluate McG on the CLEVR and VQA-Abstract datasets. Extensive experiments show that McG has achieved the state-of-the-art performance on these benchmark datasets. [ABSTRACT FROM AUTHOR]

The recent rise in artificial intelligence systems, such as ChatGPT, poses a fundamental problem for the educational sector. In universities and schools, many forms of assessment, such as coursework, are completed without invigilation. Therefore, students could hand in work as their own which is in fact completed by AI. Since the COVID pandemic, the sector has additionally accelerated its reliance on unsupervised 'take home exams'. If students cheat using AI and this is undetected, the integrity of the way in which students are assessed is threatened. We report a rigorous, blind study in which we injected 100% AI written submissions into the examinations system in five undergraduate modules, across all years of study, for a BSc degree in Psychology at a reputable UK university. We found that 94% of our AI submissions were undetected. The grades awarded to our AI submissions were on average half a grade boundary higher than that achieved by real students. Across modules there was an 83.4% chance that the AI submissions on a module would outperform a random selection of the same number of real student submissions. [ABSTRACT FROM AUTHOR]

With the advent of ChatGPT and other large language models, the number of artificial texts we encounter on a daily basis is about to increase substantially. This essay asks how this new textual situation may influence what one can call the "standard expectation of unknown texts," which has always included the assumption that any text is the work of a human being. As more and more artificial writing begins to circulate, the essay argues, this standard expectation will shift—first, from the immediate assumption of human authorship to, second, a creeping doubt: did a machine write this? In the wake of what Matthew Kirschenbaum has called the "textpocalypse," however, this state cannot be permanent. The author suggests that after this second transitional period, one may suspend the question of origins and, third, take on a post-artificial stance. One would then focus only on what a text says, not on who wrote it; post-artificial writing would be read with an agnostic attitude about its origins. This essay explores the implications of such post-artificiality by looking back to the early days of text synthesis, considering the limitations of aesthetic Turing tests, and indulging in reasoned speculation about the future of literary and nonliterary text generation. [ABSTRACT FROM AUTHOR]

The question of what generates conscious experience has mesmerized thinkers since the dawn of humanity, yet its origins remain a mystery. The topic of consciousness has gained traction in recent years, thanks to the development of large language models that now arguably pass the Turing test, an operational test for intelligence. However, intelligence and consciousness are not related in obvious ways, as anyone who suffers from a bad toothache can attest—pain generates intense feelings and absorbs all our conscious awareness, yet nothing particularly intelligent is going on. In the hard sciences, this topic is frequently met with skepticism because, to date, no protocol to measure the content or intensity of conscious experiences in an observer-independent manner has been agreed upon. Here, we present a novel proposal: Conscious experience arises whenever a quantum mechanical superposition forms. Our proposal has several implications: First, it suggests that the structure of the superposition determines the qualia of the experience. Second, quantum entanglement naturally solves the binding problem, ensuring the unity of phenomenal experience. Finally, a moment of agency may coincide with the formation of a superposition state. We outline a research program to experimentally test our conjecture via a sequence of quantum biology experiments. Applying these ideas opens up the possibility of expanding human conscious experience through brain–quantum computer interfaces. [ABSTRACT FROM AUTHOR]

Field of Light Displays (FoLDs) promise to allow further increases in realism beyond the limits of conventional 2D displays. While the question of how a conventional display can match the limits of the human visual system has been answered, the same question applied to FoLDs has been given much less attention. In this work, we review a recent acuity‐limited viewer model of resolution at depth for a FoLD display. We validate this model using display simulation. We discuss how this model gives less conservative resolution guidelines than previous models, presenting resolution requirements for passing the visual Turing test. [ABSTRACT FROM AUTHOR]

Simulation is a powerful approach that plays a significant role in science and technology. Computational models that simulate learner interactions and data hold great promise for educational technology as well. Amongst others, simulated learners can be used for teacher training, for generating and evaluating hypotheses on human learning, for developing adaptive learning algorithms, for building virtual worlds in which students can practice collaboration skills with simulated pals, and for testing learning environments. This paper provides the first systematic literature review on simulated learners in the broad area of artificial intelligence in education and related fields, focusing on the decade 2010-19. We analyze the trends regarding the use of simulated learners in educational technology within this decade, the purposes for which simulated learners are being used, and how the validity of the simulated learners is assessed. We find that simulated learner models tend to represent only narrow aspects of student learning. And, surprisingly, we also find that almost half of the studies using simulated learners do not provide any evidence that their modeling addresses the most fundamental question in simulation design – is the model valid? This poses a threat to the reliability of results that are based on these models. Based on our findings, we propose that future research should focus on developing more complete simulated learner models. To validate these models, we suggest a standard and universal criterion, which is based on the lasting idea of Turing's Test. We discuss the properties of this test and its potential to move the field of simulated learners forward. [ABSTRACT FROM AUTHOR]

"Sounding Human: Music and Machines, 1740–2020" by Deirdre Loughridge explores the historical and evolving relationship between humans and machines in the creation and perception of music. The book delves into various case studies, spanning from Enlightenment France to contemporary America, to challenge the dichotomy between human and machine music. Through analyses of automata, hybrids, analogies, personification of musical instruments, and posthumanist sensibilities in pop music, Loughridge offers a nuanced perspective on the complex interplay between humans and machines in musical expression. The book encourages historians of technology to reconsider their understanding of human-machine relations and provides a fresh examination of the intersections between technology, music, and culture. [Extracted from the article]

The release of the generative pre-trained transformer (GPT) series has brought artificial general intelligence (AGI) to the forefront of the artificial intelligence (AI) field once again. However, the questions of how to define and evaluate AGI remain unclear. This perspective article proposes that the evaluation of AGI should be rooted in dynamic embodied physical and social interactions (DEPSI). More specifically, we propose five critical characteristics to be considered as AGI benchmarks and suggest the Tong test as an AGI evaluation system. The Tong test describes a value- and ability-oriented testing system that delineates five levels of AGI milestones through a virtual environment with DEPSI, allowing for infinite task generation. We contrast the Tong test with classical AI testing systems in terms of various aspects and propose a systematic evaluation system to promote standardized, quantitative, and objective benchmarks and evaluation of AGI.

AI chatbots have recently fuelled debate regarding education practices in higher education institutions worldwide. Focusing on Generative AI and ChatGPT in particular, our study examines how AI chatbots impact university teachers' assessment practices, exploring teachers' perceptions about how ChatGPT performs in response to home examination prompts in undergraduate contexts. University teachers (n = 24) from four different departments in humanities and social sciences participated in Turing Test-inspired experiments, where they blindly assessed student and ChatGPT-written responses to home examination questions. Additionally, we conducted semi-structured interviews in focus groups with the same teachers examining their reflections about the quality of the texts they assessed. Regarding chatbot-generated texts, we found a passing rate range across the cohort (37.5 − 85.7%) and a chatbot-written suspicion range (14–23%). Regarding the student-written texts, we identified patterns of downgrading, suggesting that teachers were more critical when grading student-written texts. Drawing on post-phenomenology and mediation theory, we discuss AI chatbots as a potentially disruptive technology in higher education practices. [ABSTRACT FROM AUTHOR]

How do individuals discriminate between what is human-made and what is produced by Artificial Intelligence (AI)? Despite OpenAI's mission to ensure that AI benefits humanity, their cutting-edge technology, namely ChatGPT, an AI that aims to reproduce natural human language, raises several questions about its widespread use. This contribution aims to answer the following Research Questions: RQ1 - Are users with no specific knowledge in the field of AI able to distinguish between text produced by ChatGPT or similar language models and text produced by humans? RQ2 - Is there a significant correlation between attribution of text to AI (or human) and specific opinions and attitudes? This exploratory survey does not intend to generalise the results but to identify possible opinions and attitudes that might have influenced how the participants responded. One hundred people participated in the experiment, which consisted of a survey on their knowledge and perception of ChatGPT and a two-shot Turing Test. They were asked to read various short paragraphs and try to recognise which were written by humans and which were generated by AI. The results showed that the group analysed experienced severe difficulties in recognising whether a sentence was written by an AI or a human being, that certain perceptual biases interfere with the attribution of a trivially false text, and that the attribution error can be reduced through experience and learning. Although in need of further investigation, these findings can help lay the groundwork for the effects of the interaction between humans and AIs from a social science and computer science perspective. [ABSTRACT FROM AUTHOR]

The article explores concerns and perceptions surrounding artificial intelligence (AI) in architecture, emphasizing the limitations and advantages of AI-generated designs based on student projects. It also discusses Alan Turing's concept of AI and the Turing Test, highlighting the ongoing challenges in achieving true AI despite advancements in technology.

Advances in artificial intelligence (AI) raise important questions about whether people view moral evaluations by AI systems similarly to human-generated moral evaluations. We conducted a modified Moral Turing Test (m-MTT), inspired by Allen et al. (Exp Theor Artif Intell 352:24–28, 2004) proposal, by asking people to distinguish real human moral evaluations from those made by a popular advanced AI language model: GPT-4. A representative sample of 299 U.S. adults first rated the quality of moral evaluations when blinded to their source. Remarkably, they rated the AI's moral reasoning as superior in quality to humans' along almost all dimensions, including virtuousness, intelligence, and trustworthiness, consistent with passing what Allen and colleagues call the comparative MTT. Next, when tasked with identifying the source of each evaluation (human or computer), people performed significantly above chance levels. Although the AI did not pass this test, this was not because of its inferior moral reasoning but, potentially, its perceived superiority, among other possible explanations. The emergence of language models capable of producing moral responses perceived as superior in quality to humans' raises concerns that people may uncritically accept potentially harmful moral guidance from AI. This possibility highlights the need for safeguards around generative language models in matters of morality. [ABSTRACT FROM AUTHOR]

This article proposes the notion of the 'Lovelace Effect' as an analytical tool to identify situations in which the behaviour of computing systems is perceived by users as original and creative. It contrasts the Lovelace Effect with the more commonly known 'Lovelace objection', which claims that computers cannot originate or create anything, but only do what their programmers instruct them to do. By analysing the case study of AICAN – an AI art-generating system – we argue for the need for approaches in computational creativity to shift focus from what computers are able to do in ontological terms to the perceptions of human users who enter into interactions with them. The case study illuminates how the Lovelace effect can be facilitated through technical but also through representational means, such as the situations and cultural contexts in which users are invited to interact with the AI. [ABSTRACT FROM AUTHOR]

This paper aims, using some examples from Artificial Intelligence research, to show that passing the Turing test depends more on the cognitive characteristics of the test experimenters than on the machines subjected to the test. Furthermore, it aims to show that simulators that pass the Turing test will always have a certain degree of indeterminacy, which raises ethical questions about the purpose of building such simulators. [ABSTRACT FROM AUTHOR]

This essay shares a series of intuitions about certain paradoxes that artificial intelligence reveals when confronted with the Turing test. Using chess as an example, we ask about the feasibility of distinguishing intelligent behaviour from the ability to simulate, or even the impossibility of discrimination by the average human. We offer sense-making and corporeity, as opposed to mere computations, as the central attribute of living beings. And in attempting to discern the limits of these simulations, we even consult ChatGPT's own opinion. [ABSTRACT FROM AUTHOR]

The manual recording of electronic health records (EHRs) by clinicians in the emergency department (ED) is time-consuming and challenging. In light of recent advancements in large language models (LLMs) such as GPT and BERT, this study aimed to design and validate LLMs for automatic clinical diagnoses. The models were designed to identify 12 medical symptoms and 2 patient histories from simulated clinician–patient conversations within 6 primary symptom scenarios in emergency triage rooms. We developed classification models by fine-tuning BERT, a transformer-based pre-trained model. We subsequently analyzed these models using eXplainable artificial intelligence (XAI) and the Shapley additive explanation (SHAP) method. A Turing test was conducted to ascertain the reliability of the XAI results by comparing them to the outcomes of tasks performed and explained by medical workers. An emergency medicine specialist assessed the results of both XAI and the medical workers. We fine-tuned four pre-trained LLMs and compared their classification performance. The KLUE-RoBERTa-based model demonstrated the highest performance (F1-score: 0.965, AUROC: 0.893) on human-transcribed script data. The XAI results using SHAP showed an average Jaccard similarity of 0.722 when compared with explanations of medical workers for 15 samples. The Turing test results revealed a small 6% gap, with XAI and medical workers receiving the mean scores of 3.327 and 3.52, respectively. This paper highlights the potential of LLMs for automatic EHR recording in Korean EDs. The KLUE-RoBERTa-based model demonstrated superior classification performance. Furthermore, XAI using SHAP provided reliable explanations for model outputs. The reliability of these explanations was confirmed by a Turing test. • The data was collected from simulated clinician-patient conversations. • The fine-tuned large language model identifies medical information included in electronic health records. • The outcomes of the model were interpreted through eXplainable AI. • The Turing test was conducted to demonstrate the reliability of the eXplainable AI results. [ABSTRACT FROM AUTHOR]

In designing an intelligent tutoring system, a core area of the application of AI in education, tips from the system or virtual tutors are crucial in helping students solve difficult questions in disciplines like mathematics. Traditionally, the manual design of general tips by teachers is time-consuming and error-prone. Generative AI, like ChatGPT, presents a new channel for designing general tips. This study utilized prompt engineering and Chain of Thought to summarize general tips for given mathematical problems (one geometry problem and one algebra problem) and their solutions. A Turing test was conducted to compare ChatGPT-generated general tips with human-designed ones. Results from 121 human evaluators, each assessing 6 ChatGPT-generated and 6 human-designed general tips for each of two mathematical problems, showed that the average score for ChatGPT-generated tips is less than that of human-designed tips at a statistically significant level (p < 0.05), and Zero-Shot CoT achieved the best score. However, no evaluator could distinguish the tip types exactly. The average precision, recall and F-value of all ChatGPT-generated tips are less than 40%. AI-generated general tips can serve as a valuable reference for teachers to enhance efficiency and students' mathematical learning. [ABSTRACT FROM AUTHOR]

Copyright of Felsefe ve Sosyal Bilimler Dergisi (FLSF) is the property of Felsefe ve Sosyal Bilimler Dergisi (FLSF) and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Artificial Intelligence (AI) and related technologies have a long history of being used in education for motivating learners and enhancing learning. However, there have also been critiques for a too uncritical and naïve implementation of AI in education (AIED) and the potential misuse of the technology. With the release of the virtual assistant ChatGPT from OpenAI, many educators and stakeholders were both amazed and horrified by the potential consequences for education. One field with a potential high impact of ChatGPT is programming education in Computer Science (CS), where creating assessments has long been a challenging task due to the vast amount of programming solutions and support on the Internet. This now appears to have been made even more challenging with ChatGPT's ability to produce both complex and seemingly novel solutions to programming questions. With the support of data collected from interactions with ChatGPT during the spring semester of 2023, this position paper investigates the potential opportunities and threats of ChatGPT for programming education, guided by the question: What could the potential consequences of ChatGPT be for programming education? This paper applies a methodological approach inspired by analytic autoethnography to investigate, experiment, and understand a novel technology through personal experiences. Through this approach, the authors have documented their interactions with ChatGPT in field diaries during the spring semester of 2023. Topics for the questions have related to content and assessment in higher education programming courses. A total of 6 field diaries, with 82 interactions (1 interaction = 1 question + 1 answer) and additional reflection notes, have been collected and analysed with thematic analysis. The study finds that there are several opportunities and threats of ChatGPT for programming education. Some are to be expected, such as that the quality of the question and the details provided highly impact the quality of the answer. However, other findings were unexpected, such as that ChatGPT appears to be "lying" in some answers and to an extent passes the Turing test, although the intelligence of ChatGPT should be questioned. The conclusion of the study is that ChatGPT have potential for a significant impact on higher education programming courses, and probably on education in general. The technology seems to facilitate both cheating and enhanced learning. What will it be? Cheating or AI-enhanced learning? This will be decided by our actions now since the technology is already here and expanding fast. [ABSTRACT FROM AUTHOR]

This article is a response to the position of Landgrebe and Smith on the fundamental limitations that prevent the creation of general artificial intelligence (AGI), expressed in their book Why Machines Will Never Rule the World. The reasons for failures for attempts to create AGI using formal logic and algorithmic approaches to modeling intelligence are discussed. An attempt is made to define the future direction of intellectual systems development as hybrid evolving systems, as well as a revision of the Turing test statement and language models role. [ABSTRACT FROM AUTHOR]

The integration of Artificial Intelligence (AI) in the media results in the publication of thousands of automated news articles in Spanish every day. This study uses a Turing test to compare the quality of news articles written by professional journalists (from Efe) with those produced by natural language generation (NLG) software (from Narrativa). Based on Sundar’s dimensions (1999) crucial to news perception – credibility, readability and journalistic expertise – , an internationally validated experimental methodology is employed, exploring a novel topic in Spanish: health information. The experiment deliberately varied real and declared authorships – AI and human journalists – to detect potential biases in assessing authorship credibility. A self-administered questionnaire adapted for online surveys was used (N=222), and gender imbalances were minimized to ensure gender equality in the sample (N=128). The study reveals that there are no significant differences between news articles generated by the AI and those written by professional journalists. Both types of news are considered equally credible, though some biases are detected in the evaluation of declared authorship: the AI author is perceived as more believable than the human, while the human journalist is perceived as creating a more lively narrative. The study concludes that it is feasible to produce automated news in Spanish without compromising its quality. In the global media landscape, automated systems employing NLG, machine learning and sophisticated databases successfully advance into new domains such as health information.