Turing Test Research Articles

O-251 Generative artificial intelligence substantially enhances the accuracy of embryo selection models

Abstract Study question Can generative artificial intelligence (AI) produce high-fidelity human embryo images to improve AI-based embryo selection models? Summary answer Generative AI models exhibit the capability to generate high-fidelity human embryo images and can substantially improve the performance of AI models through extensive training data. What is known already The integration of AI into in vitro fertilization (IVF) procedures holds the potential to enhance objectivity and automate embryo selection for transfer. However, the effectiveness of AI is limited by data scarcity and ethical concerns related to patient data privacy. Generative Adversarial Networks (GAN) have emerged as a promising approach to alleviate data limitations by generating synthetic data that closely approximate real images. However, the advantage of applying GAN in IVF embryo images is yet to be determined. Study design, size, duration Embryo images were retrieved as training data from time-lapse microscopy (TLM) videos (n = 328). Embryo key morphokinetic variables including blastomere division from 1-cell stage to 9 (and more) cells-stage, compaction of morula, and blastocyst formation were manually annotated by six embryologists. A style-based GAN was fine-tuned as the generative model and a residual network as the morphokinetic classification model. Participants/materials, setting, methods We configured generative models with data augmentation (AUG) and pretrained weights (Pretrained-T/R) to test their refinement to synthetic image quality. We analyzed quantitative metrics including Fréchet Inception Distance (FID) and Kernel Inception Distance (KID) to assess the quality and fidelity of the generated images. Subsequently, we evaluated qualitative performance through a visual Turing test. Then, we compared the performance of a classifier by training real images alone and by integrating real images with generated data. Main results and the role of chance During the training process, we observed consistent improvement in image quality that was measured by FID and KID scores. The AUG+Pretrained-R model showed the highest performance of the evaluated five configurations with FID and KID scores of 15.2 and 0.004 following 5,000 training iterations. Subsequently, we carried out a visual Turing test, such that a total of 60 individuals including IVF embryologists (Group I, n = 25), IVF laboratory technicians (Group II, n = 15), and non-experts (Group III, n = 20) evaluated the synthetic blastocyst-stage embryo images. Group I displayed accuracy of 55.7% (±6.2), sensitivity of 65.9% (±14.1), and specificity of 45.1% (±17.8). Group II showed accuracy of 54.2% (±4.7), sensitivity of 65.9% (±17.6), and specificity of 42.0% (±19.6). Group III had the accuracy of 50.1% (±3.8), sensitivity of 49.3% (±6.3), and specificity was 50.9% (±10.4). Statistical analysis indicated significant differences in accuracy (Kruskal-Wallis test, P=1.2x10-3) and sensitivity (Kruskal-Wallis test, P=1.6x10-4) among the three groups. In the embryonic stage classification model, integrated with generated data outperformed the model using real data solely, with an increase of validation accuracy from 72.6% to 90.6%. The classifier was tested on external TLM video data (n = 100) where the area under the curve was 0.92. Limitations, reasons for caution Further research is needed to involve a more extensive and diverse dataset, e.g. video generation, and more detailed annotations including polar body appearance, pronuclei appearance and disappearance, and blastocyst expansion and hatching. Wider implications of the findings Generative AI offers potential in generating high-fidelity human embryo images. This approach ensures sufficient training datasets while addressing class imbalances for robust AI-based methods, ultimately transforming embryo selection and enhancing IVF outcomes. A set of 5,000 generated images (1024x1024-pixel) at each embryonic developmental stage was open-source for future AI studies. Trial registration number not applicable

P-182 High-resolution synthetic embryo images generated by AI

Abstract Study question How effectively can a latent diffusion model generate high-fidelity embryo images tailored to the specific contextual needs of researchers, based on the user’s text input? Summary answer The latent diffusion model successfully generated high-resolution embryo images, providing a novel approach to address data scarcity in embryo imaging. What is known already The use of AI in generating synthetic data has been increasingly explored to address the scarcity of real-world datasets in various fields. Particularly in medical imaging, AI-generated synthetic data offers a potential solution to overcome the limitations imposed by data privacy concerns and ethical considerations. Previous studies have shown that AI can replicate complex patterns in data, but its application in generating embryo images remains less explored. Study design, size, duration Single static images of 5,133 Day 5 blastocysts and 2,093 Day 3 cleavages were retrospectively collected from seven in vitro fertilization clinics between June 2011 and May 2022. The images were analyzed along with relevant metadata including clinical information and embryo grades. Day 3 embryo grading was based on the number of blastomeres, evenness, and fragmentation percentage. Day 5 grading criteria included the inner cell mass, trophectoderm, and blastocyst stage evaluations. Participants/materials, setting, methods An AI model using latent diffusion was developed to generate 10,051 Day 5 and 10,088 Day 3 synthetic embryo images. The authenticity was assessed through visual Turing tests, where embryologists discerned real from synthetic images. For the evaluation, 200 real (100 Day 5, 100 Day 3) and 200 synthetic (100 Day 5, 100 Day 3) images were randomly chosen from each dataset, ensuring a comprehensive test of the generated images’ realism. Main results and the role of chance The AI model’s efficacy in generating synthetic embryo images was assessed through visual Turing tests on Day 5 and Day 3 embryos, yielding accuracies of 0.59 and 0.57, respectively. These accuracies fall within the 99% confidence interval of near-random performance (0.41 to 0.59), highlighting the challenge in distinguishing synthetic from real images. For Day 5 embryos, the sensitivity and specificity were recorded at 0.52 and 0.66, indicating a moderate challenge in identifying synthetic images and a relatively higher ease in recognizing real ones. Day 3 embryos presented a lower sensitivity of 0.41, suggesting greater difficulty in detecting synthetic images, while the specificity increased to 0.73, indicating a stronger ability to identify real images. Collectively, with an overall accuracy of 0.58, sensitivity of 0.47, and specificity of 0.70, these findings confirm the synthetic images’ remarkable realism, closely emulating actual embryo features to the extent that differentiation by experts proved challenging. This level of realism underscores the synthetic images’ potential to significantly enrich embryological research datasets, promising advancements in the field. Limitations, reasons for caution The study’s limitation lies in the generated synthetic embryos being based on three embryo grade categories, limiting feature diversity. Additionally, there was no quantitative assessment method for these images, relying instead on expert evaluation, which required deploying a large number of embryologists. Wider implications of the findings The model’s generation of highly realistic embryo images counters data scarcity in embryological research, potentially elevating AI’s utility, enriching educational content, advancing reproductive medicine, and ensuring ethical data usage. Trial registration number not applicable

Stepwise Transfer Learning for Expert-level Pediatric Brain Tumor MRI Segmentation in a Limited Data Scenario.

Purpose To develop, externally test, and evaluate clinical acceptability of a deep learning pediatric brain tumor segmentation model using stepwise transfer learning. Materials and Methods In this retrospective study, the authors leveraged two T2-weighted MRI datasets (May 2001 through December 2015) from a national brain tumor consortium (n = 184; median age, 7 years [range, 1-23 years]; 94 male patients) and a pediatric cancer center (n = 100; median age, 8 years [range, 1-19 years]; 47 male patients) to develop and evaluate deep learning neural networks for pediatric low-grade glioma segmentation using a stepwise transfer learning approach to maximize performance in a limited data scenario. The best model was externally tested on an independent test set and subjected to randomized blinded evaluation by three clinicians, wherein they assessed clinical acceptability of expert- and artificial intelligence (AI)-generated segmentations via 10-point Likert scales and Turing tests. Results The best AI model used in-domain stepwise transfer learning (median Dice score coefficient, 0.88 [IQR, 0.72-0.91] vs 0.812 [IQR, 0.56-0.89] for baseline model; P = .049). With external testing, the AI model yielded excellent accuracy using reference standards from three clinical experts (median Dice similarity coefficients: expert 1, 0.83 [IQR, 0.75-0.90]; expert 2, 0.81 [IQR, 0.70-0.89]; expert 3, 0.81 [IQR, 0.68-0.88]; mean accuracy, 0.82). For clinical benchmarking (n = 100 scans), experts rated AI-based segmentations higher on average compared with other experts (median Likert score, 9 [IQR, 7-9] vs 7 [IQR 7-9]) and rated more AI segmentations as clinically acceptable (80.2% vs 65.4%). Experts correctly predicted the origin of AI segmentations in an average of 26.0% of cases. Conclusion Stepwise transfer learning enabled expert-level automated pediatric brain tumor autosegmentation and volumetric measurement with a high level of clinical acceptability. Keywords: Stepwise Transfer Learning, Pediatric Brain Tumors, MRI Segmentation, Deep Learning Supplemental material is available for this article. © RSNA, 2024.

Beyond the Turing Test: Exploring the implications of generative AI for category construction

As generative AI systems move beyond Turing’s benchmark for whether a machine exhibits human-like intelligence, what implications does this technological milestone have for organization theory? We engage with this question by considering how the increasing creativity and social competence exhibited by generative AI impacts processes of social construction and cultural evolution that have, up to this point, been the exclusive domain of humans. More specifically, we consider what it means to have intelligent machines capable of category work, which we define here as both the culturally savvy use of categories and purposeful participation in the processes of construction that underpin systems of categories more generally. We go on to explore some of the implications for individuals, organizations and societies of the appearance of this new class of artificial participants in the processes that constitute category systems.

Boosting the transferability of adversarial CAPTCHAs

Completely Automated Public Turing test to tell Computers and Humans Apart (CAPTCHA) is a test to distinguish humans and computers. Since attackers can achieve high accuracy in recognizing the CAPTCHAs using deep learning models, geometric transformations are added to the CAPTCHAs to disturb deep learning model recognition. However, excessive geometric transformations might also affect humans’ recognition of the CAPTCHA. Adversarial CAPTCHAs are special CAPTCHAs that can disrupt deep learning models without affecting humans. Previous works of adversarial CAPTCHAs mainly focus on defending the filtering attack. In real-world scenarios, the attackers’ models are inaccessible when generating adversarial CAPTCHAs, and the attackers may use models with different architectures, thus it is crucial to improve the transferability of the adversarial CAPTCHAs. We propose CFA, a method to generate more transferable adversarial CAPTCHAs focusing on altering content features in the original CAPTCHA. We use the attack success rate as our metric to evaluate the effectiveness of our method when attacking various models. A higher attack success rate means a higher level of preventing models from recognizing the CAPTCHAs. The experiment shows that our method can effectively attack various models, even when facing possible defense methods that the attacker might use. Our method outperforms other feature space attacks and provides a more secure version of adversarial CAPTCHAs.

A real-world test of artificial intelligence infiltration of a university examinations system: A "Turing Test" case study.

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.

ImageVeriBypasser: An image verification code recognition approach based on Convolutional Neural Network

AbstractThe 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.

Beyond Turing: Testing LLMs for Intelligence

Modern large language models appear to be able to fool the Turing Test. Will we know if they are becoming as intelligent as people?

Novel AI-based automated virtual implant placement: Artificial versus human intelligence

ObjectivesTo assess quality, clinical acceptance, time-efficiency, and consistency of a novel artificial intelligence (AI)-driven tool for automated presurgical implant planning for single tooth replacement, compared to a human intelligence (HI)-based approach. Materials and methodsTo validate a novel AI-driven implant placement tool, a dataset of 10 time-matching cone beam computed tomography (CBCT) scans and intra-oral scans (IOS) previously acquired for single mandibular molar/premolar implant placement was included. An AI pre-trained model for implant planning was compared to human expert-based planning, followed by the export, evaluation and comparison of two generic implants—AI-generated and human-generated—for each case. The quality of both approaches was assessed by 12 calibrated dentists through blinded observations using a visual analogue scale (VAS), while clinical acceptance was evaluated through an AI versus HI battle (Turing test). Subsequently, time efficiency and consistency were evaluated and compared between both planning methods. ResultsOverall, 360 observations were gathered, with 240 dedicated to VAS, of which 95 % (AI) and 96 % (HI) required no major, clinically relevant corrections. In the AI versus HI Turing test (120 observations), 4 cases had matching judgments for AI and HI, with AI favoured in 3 and HI in 3. Additionally, AI completed planning more than twice as fast as HI, taking only 198 ± 33 s compared to 435 ± 92 s (p < 0.05). Furthermore, AI demonstrated higher consistency with zero-degree median surface deviation (MSD) compared to HI (MSD=0.3 ± 0.17 mm). ConclusionAI demonstrated expert-quality and clinically acceptable single-implant planning, proving to be more time-efficient and consistent than the HI-based approach. Clinical significancePresurgical implant planning often requires multidisciplinary collaboration between highly experienced specialists, which can be complex, cumbersome and time-consuming. However, AI-driven implant planning has the potential to allow clinically acceptable planning, significantly more time-efficient and consistent than the human expert.

Simulating cross‐modal medical images using multi‐task adversarial learning of a deep convolutional neural network

AbstractComputed 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 36.81 4.46 HU, peak signal to noise ratio of 26.12 0.31 dB, and structural similarity measure of 0.9 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.

Intermedial and theatrical perspectives of AI: Re-framing the Turing test

Artificial intelligence (AI), the understanding and building of computational agents that act intelligently, is claimed to be a powerful, pervasive medium. Although we interact with it and hear a lot about it, we do not ‘see’ AI but experience its manifestations and/or outcomes, such as chatterbots and virtual assistants. A criterion for deeming an artificial agent as intelligent has been already proposed by Turing in 1950, coined as the ‘Imitation Game’, where a machine takes the place of a man, known as the Turing test. Despite the test being initially conceived as a make-believe game, AI has been enmeshed in major fields of human social activity and co-determines our lives. In this article, AI is compared with the media type of theatre performance, the epitome of make-believe, on the basis of intermediality. Furthermore, the analogies between AI and theatre are discussed and the paradigm of the puppet theatre as well as the medium of the mask prevail in the analysis. Findings are discussed, especially in light of the mind–body split and the alignment problem, and their implications are contemplated, allowing a re-estimation and re-framing of the Turing test in its theatrical and performative dimension.

Tender Flesh: Theorizing the Hybrid Embodied Poetics of the Asian American Queer Femme in Franny Choi's Soft Science

Abstract: This essay investigates how the grotesque and transgressive aesthetic of Franny Choi's Soft Science (2019) reimagines the Asian American Queer Femme subject's relationship to language, the body, and consciousness. Choi's work unravels this relationship by destabilizing human/nonhuman binary through a polyphonic poetic voice. This act of shapeshifting between humans, cyborgs, cephalopods, and other forms functions as a critique of the oppressive power structures of white supremacy, patriarchy, and cisheteronormativity. To theorize her poetics, I foreground the critical lenses of Donna Haraway's posthuman feminism, Lisa Lowe's Asian American aesthetics, Vivian Huang's inscrutability and Mel Chen's queer linguistic animacies in connection with each other as a framework for understanding the hybrid embodiment of the Asian American Queer Femme. Subsequently, I trace Choi's "hybrid embodied poetics"—a hermeneutic that emphasizes her multitudinous lyric voice—primarily within the poems "Glossary of Terms," "Turing Test," "Shokushu Goukan for the Cyborg Soul," "Chi." Furthermore, this essay emphasizes the interrelated nature of the Asian American Queer Femme's embodied experiences and identities, as Choi's work repositions the theoretical frames of Haraway, Lowe, Huang, and Chen in an intersectional dialogue with each other.

P‐135: Towards Passing the Visual Turing Test with Field of Light Displays

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.

On Artificial and Post-artificial Texts: Machine Learning and the Reader's Expectations of Literary and Non-literary Writing

Abstract 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.

Testing the Conjecture That Quantum Processes Create Conscious Experience.

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.

Clinical Turing tests with user certainty analysis to create and validate synthetic electrocardiogram images for artificial intelligence-enhanced algorithm development

Abstract Background Artificial intelligence-enhanced electrocardiogram (AI-ECG) algorithms have primarily been created using digitised signal data, owing to a relative absence of publicly available image-based datasets. ECGs are often scanned or photographed into electronic health records. For maximum clinical utility, AI-ECG algorithms should be applicable to these data. Synthetic data could expedite the creation of extensive, fully anonymised image-based ECG datasets to permit training image-based AI algorithms, but it is essential that such datasets contain the artefacts encountered in clinical practice. We investigated whether iterative clinical Turing tests with user certainty analysis could be used to develop and validate synthetic ECG data. Purpose To create synthetic ECG images containing the artefacts typically encountered in clinical practice, and to validate the images through iterative Turing testing and user certainty analysis. Methods Synthetic ECG images containing artefacts were created using the PTB-XL dataset (a publicly available signal-based dataset comprising 21799 ECGs) as source data. Iterative clinical Turing tests were conducted where healthcare professionals completed an online survey comprising 60 real and synthetic ECGs. Participants were asked to select whether they thought ECGs were real or synthetic. For user certainty analysis, participants were asked to rate their confidence in their answers using a five-point Likert scale (Figure 1). Likert scale responses were converted into a signed ordinal scale representing user certainty in the identification of real or synthetic data. This scale was used to perform Receiver Operating Characteristic (ROC) analysis. Following quantitative survey completion, qualitative feedback was sought and used to iteratively improve the realism of the synthetic images. Results A total of 26 healthcare professionals completed the clinical Turing tests over three rounds. Qualitative feedback was used to improve the fidelity of the synthetic ECG images between rounds (Table 1). During iterative testing, the proportion of synthetic ECGs correctly identified fell from 61.5% to 53.7%, and the proportion of real-world ECGs correctly identified fell from 66.3% to 53.0% (Figure 1). Following the final Turing test, ROC analysis revealed no discriminative ability for identifying synthetic data (C-statistic 0.480, 95% confidence interval 0.432-0.529). Conclusion Iterative Turing testing with user certainty analysis and qualitative user feedback may be used to create synthetic ECG images containing the artefacts typically encountered in clinical practice. Iterative Turing testing improved the images’ realism confirming their potential to augment image-based AI algorithm development. The presented methodology establishes a framework to develop high fidelity, synthetic patient datasets presenting a significant opportunity to enhance the uptake of AI within electrophysiology, cardiology, and medicine.

ТРЕНУВАННЯ НЕЙРОМЕРЕЖІ ДЛЯ СТИЛІЗАЦІЇ ЗОБРАЖЕНЬ

Improving the visualization of projects and portfolios of designers and architects can be achieved by enhancing the illustrativeness and stylization of images using artificial intelligence. The use of neural networks for content generation significantly speeds up the work of designers. Among all the neural networks for image generation, MidJourney shows the best results. After analyzing the licenses and subscription costs of services as well as the models they employ, the Stable Diffusion deep learning neural network was chosen. The Stable Diffusion neural network is open-source, unlike DALL-E and Midjourney, allowing for unlimited content generation. The stylized images were generated in Stable Diffusion using Dreambooth based on the Google Collab platform. The creation of a custom model was conducted in two stages. The first stage involved preparing images for training the Stable Diffusion neural network. The second stage was the direct training of the neural network based on the Google Collab platform. Kobzar's graphic drawings served as the training dataset. Initially, 77 drawings with the same theme were selected for model training. 30 of these were used to train the model after corrections in Adobe Photoshop and Topaz Photo AI. Adjustments included cropping, background removal, printing raster, noise reduction, sharpening, and scaling images. The originality of the work lies in the fact that the trained model was used to create stylized creative images, utilizing excerpts from the poet's poems describing nature and events in a very realistic way. The generated images have successfully passed the Turing test, indicating a realistic reproduction of the style of Taras Shevchenko's drawings and the utilization of the author's poetic text as a prompt. The use of neural networks for generating and styling images as virtual assistants for designers and architects speeds up the creative process and enables the creation of works of any complexity.

Overview, Concepts, Viewpoints to the Possibility of Legal Regulation of Artificial Intelligence Technology

The purpose of the study is to disclose the concept of "artificial intelligence", to outline the possibilities of legal regulation of AI. The article considers the studies of various authors, among which it is necessary to name P.M. Morhat, A. Romaglio, V.L. Entin. The scientific novelty of the work lies in the fact that the category of "artificial intelligence" is considered from the position of the legislation of the Russian Federation, as well as based on the opinion of domestic and foreign scientists. As a result, based on the comparative analysis of information contained in the works of various authors, regulatory legal acts of the Russian Federation, foreign judicial practice, it is concluded that representatives of the legal and scientific communities demonstrate the lack of consensus on the question of whether the units of artificial intelligence and neural networks can be subjects of copyright. A. Ramalho is convinced that AI units cannot be subjects of copyright, which is confirmed by the analysis of the following court decisions: Decision of the Court of Appeal of the Federal Court of Australia dated 13.04.2022, Decision of Shenzhen Nanshan District People's Court dated 24 December 2019. The opposing view is reflected in the Japanese "Intellectual Property Strategy Program 2016", which uses the following argument: it seems impossible to distinguish the product of AI units from human intellectual products. It is impossible to agree with this, as the images generated by neural networks often cause the "uncanny valley" effect in people; no neural network has been able to pass the Turing test so far.

Approach to Detecting Malicious Bots in the Vkontakte Social Network and Assessing Their Parameters

The emergence of new varieties of bots in social networks and the improvement of their capabilities to imitate the natural behavior of real users represent a significant problem in the field of protection of social networks and online communities. This paper proposes a new approach to detecting and assessing the parameters of bots within the social network «VKontakte». The basis of the proposed approach is the creation of datasets using the method of «controlled purchase» of bots, which allows one to assess bots’ characteristics such as price, quality, and speed of action of bots, and using the Turing Test to assess how much users trust bots. In combination with traditional machine learning methods and features extracted from interaction graphs, text messages, and statistical distributions, it becomes possible to not only detect bots accurately but also predict their characteristics. This paper demonstrates that the trained machine learning model, based on the proposed approach, is robust to imbalanced data and can identify most types of bots as it has only a minor correlation with their main characteristics. The proposed approach can be used within the choice of countermeasures for the protection of social networks and for historical analysis, which allows not only to confirm the presence of bots but also to characterize the specifics of the attack.

Attributions toward artificial agents in a modified Moral Turing Test

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.