Search

We discuss the emerging new opportunity for building feedback-rich computational models of social systems using generative artificial intelligence. Referred to as Generative Agent-Based Models (GABMs), such individual-level models utilize large language models to represent human decision-making in social settings. We provide a GABM case in which human behavior can be incorporated in simulation models by coupling a mechanistic model of human interactions with a pre-trained large language model. This is achieved by introducing a simple GABM of social norm diffusion in an organization. For educational purposes, the model is intentionally kept simple. We examine a wide range of scenarios and the sensitivity of the results to several changes in the prompt. We hope the article and the model serve as a guide for building useful dynamic models of various social systems that include realistic human reasoning and decision-making.

In the context of the advancement of software architectures based on microservices and the relevance of requirements engineering in application development, the problem of manual creation and the need for expertise to define microservices, a human activity that requires a high level of knowledge and experience, arises. To address this challenge, a solution is proposed that consists of finding the ideal names for microservices. To automate the process of identifying microservice names, this proposal is based on the application of Natural Language Processing (NLP) techniques, graph analysis and community detection, including the use of artificial intelligence language models such as ChatGPT.

Large language models (LLMs) are being criticized for copyright infringement, inadvertent bias in training data, a danger to human innovation, the possibility of distributing incorrect or misleading information, and prejudice. Due to their popularity among students, the introduction of many comparable apps, and the inability to resist unfair and fraudulent student usage, their educational use needs to be adapted and harmonized. The incorporation of LLMs should be defined not only by pedagogues and educational institutions, but also by students who will actively utilize them to learn and prepare assignments. In order to find out what students from two universities think and suggest about LLMs use in education, they were asked to give their contribution by answering the survey that was conducted at the beginning of the spring semester of academic 2022/23. Their feedback was quantitatively and qualitatively analyzed, showing in a better light what students think about LLMs and how and why they would use them. Based on the analysis, the authors propose an original strategy for integrating LLMs into education. The proposed approach is also adapted for those students who are not interested in using LLMs and for those who prefer the hybrid mode by combining their own research with LLMs generated recommendations. The authors expect that by implementing the proposed strategy, schools will benefit from a better education in which research, creativity, academic honesty, recognition of false information, and the ability to improve knowledge will prevail.

This research sought to identify changes in terms of word choice, sentence structure, and semantics introduced by ChatGPT and how the functionality of ChatGPT impacts the pragmatics of the English language: politeness, deixis, and discourse markers. This study used corpus analysis as part of the methodology involving Natural Language Processing (NLP) tools for part-of-speech tagging and sentiments, analyzing changes in wordings, syntaxes, and semantic elements generated by ChatGPT as an advanced language modal. A qualitative approach, including user surveys and comparative analysis, proved useful in clarifying the changing patterns of politeness, deixis, and discourse markers in AI-supported communication. Henceforth, ChatGPT generates neologism, unique cultural references, verbal contextual adaptations, and a wide range of technical terms. Moreover, the researchers found that user preferences for communication mediums vary, which showed a complex field of users and their expectations in the context of language generated by ChatGPT. Thus, it was observed that ChatGPT contributes significantly to developing the pragmatic aspects of language use in the modern era. Hence, the study proposes that the ChatGPT polishes the usage of politeness terms and improves its deixis along with formal discourse markers. This highlights the need for continuous improvement, especially in areas such as politeness, deixis management, and integration of discourse markers for the developers.

This study explores the potential of ChatGPT, a large language model, in scientometrics by assessing its ability to predict citation counts, Mendeley readers, and social media engagement. In this study, 2222 abstracts from PLOS ONE articles published during the initial months of 2022 were analyzed using ChatGPT-4, which used a set of 60 criteria to assess each abstract. Using a principal component analysis, three components were identified: Quality and Reliability, Accessibility and Understandability, and Novelty and Engagement. The Accessibility and Understandability of the abstracts correlated with higher Mendeley readership, while Novelty and Engagement and Accessibility and Understandability were linked to citation counts (Dimensions, Scopus, Google Scholar) and social media attention. Quality and Reliability showed minimal correlation with citation and altmetrics outcomes. Finally, it was found that the predictive correlations of ChatGPT-based assessments surpassed traditional readability metrics. The findings highlight the potential of large language models in scientometrics and possibly pave the way for AI-assisted peer review., Human-Robot Interaction

This article examines the idea of mind-reading technology by focusing on an interesting case of applying a large language model (LLM) to brain data. On the face of it, experimental results appear to show that it is possible to reconstruct mental contents directly from brain data by processing via a chatGPT-like LLM. However, the author argues that this apparent conclusion is not warranted. Through examining how LLMs work, it is shown that they are importantly different from natural language. The former operates on the basis of nonrational data transformations based on a large textual corpus. The latter has a rational dimension, being based on reasons. Using this as a basis, it is argued that brain data does not directly reveal mental content, but can be processed to ground predictions indirectly about mental content. The author concludes that this is impressive but different in principle from technology-mediated mind reading. The applications of LLM-based brain data processing are nevertheless promising for speech rehabilitation or novel communication methods., Ethics & Philosophy of Technology

As technology progresses, there has been an increasing interest in using Chatbot GPT (Generative Pre-trained Transformer) in education. Chatbot GPT, or ChatGPT, gained one million users within the first week of launching in November 2022 and had amassed over 100 million active users by February 2023. This type of artificial intelligence uses natural language processing to convert it into a user. This paper presents a comprehensive analysis and review of 34 articles published on ChatGPT and its potential impact on education by utilizing the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) methodology. This review analyzed various studies and articles to examine the strengths and limitations of GPT language models in education from 2018 to the present. The advantages of ChatGPT include its capacity to provide personalized and adaptive learning, instant feedback, and improved accessibility. However, there are potential drawbacks, such as the lack of emotional intelligence, the risk of overreliance on technology, and privacy concerns. This review suggests that ChatGPT has significant promise for education yet reinforces the necessity for further research and careful consideration of possible risks and limitations. Specifically, it pointed out potential invisible manipulations by instructing ChatGPT to answer educationrelated topics. The paper concludes by discussing the implications of ChatGPT for the future of education and emphasizing the need for further research in this field.

Academics and educators across the world are grappling with how OpenAI’s new software, ChatGPT, will impact teaching and learning. This essay explores ChatGPT’s response to a typical calculus problem as a way of illustrating its functionality and limitations.

This study aims to analyze the Impacts of Social Environment on Interest, Critical Thinking and Academic Achievement of Students Taught Organic Chemistry Using Artificial Intelligence ChatGPT. Correlational research design was engaged for the study. 50-students in SS3 with an average age of 17years, who have access to android phones and the ChatGPT application from two senior secondary schools in Gwale Local Government area of Kano State, Nigeria formed the study sample. The instruments used for data collection include; the students’ social environment, interest and critical thinking questionnaire (r=0.96) and Organic Chemistry Achievement Test (r=0.92). The data were analyzed using mean, standard deviation and Spearman Rank Order Correlation. The study revealed a high positive correlation between the social environment and student’s interest; Social environment and critical thinking; as well as social environment and students’ achievement in the ChatGPT’s Organic Chemistry Classroom. The study further found a strong positive correlation between the students’ interest and critical thinking, critical thinking and students’ achievement in Organic Chemistry among students in the ChatGPT condition. It was concluded that social environment has a strong positive impact on students’ interest, critical thinking and academic achievement in organic chemistry among students in the ChatGPT classrooms. The study recommends that, as the artificial intelligence (AI) ChatGPT is migrating into the chemistry classrooms, government needs to re-train chemistry teachers, school librarians and lab-technologists through refreshers courses to improve their performance in the concepts

Advanced data mining methods have shown a promising capacity in building energy management. However, in the past decade, such methods are rarely applied in practice, since they highly rely on users to customize solutions according to the characteristics of target building energy systems. Hence, the major barrier is that the practical applications of such methods remain laborious. It is necessary to enable computers to have the human-like ability to solve data mining tasks. Generative pre-trained transformers (GPT) might be capable of addressing this issue, as some GPT models such as GPT-3.5 and GPT-4 have shown powerful abilities on interaction with humans, code generation, and inference with common sense and domain knowledge. This study explores the potential of the most advanced GPT model (GPT-4) in three data mining scenarios of building energy management, i.e., energy load prediction, fault diagnosis, and anomaly detection. A performance evaluation framework is proposed to verify the capabilities of GPT-4 on generating energy load prediction codes, diagnosing device faults, and detecting abnormal system operation patterns. It is demonstrated that GPT-4 can automatically solve most of the data mining tasks in this domain, which overcomes the barrier of practical applications of data mining methods in this domain. In the exploration of GPT-4, its advantages and limitations are also discussed comprehensively for revealing future research directions in this domain.

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 qual

In dem Meet-up wurde mit Teilnehmenden aus der Technikkommunikation und verschiedenen Schnittstellen sowie Expert:innen aus dem rechtlichen Bereich diskutiert, wo sich Chancen und Risiken für den Einsatz von ChatGPT im Bereich und in Schnittstellen der technischen Kommunikation ergeben. Es wurde erörtert, wie die Anwendung im Kontext der Technikkommunikation bereits umgesetzt wird und welche rechtlichen Rahmenbedingungen benötigt werden. Gemeinsames Fazit des Meet-ups war die Erkennntnis, dass rechtliche Rahmenbedingungen erst noch geschaffen werden müssen und ein achtsamer Umgang mit ChatGPT notwendig ist, damit sensible technische Daten beispielsweise von Unternehmen nicht gefährdert werden.

Large Language Models (LLMs) wie GPT-4 von OpenAI verfügen über das einzigartige Potenzial, textbasierte, komplexe Kundenanfragen automatisiert bearbeiten zu können. Sie gewährleisten dabei einen personalisierten Support auf einem standardisierten Qualitätsniveau. Es ist Ziel des Einsatzes von LLMs, eine große Anzahl von Kundeninteraktionen skalierbar zu bewältigen und damit Personalbedarf zu reduzieren. Technisch gesehen handelt es sich bei LLMs um neuronale Netze, die mit großen Mengen an Textdaten trainiert werden.

The paper aims to fulfil three main functions: (1) to serve as an introduction for the physics education community to the functioning of large language models (LLMs), (2) to present a series of illustrative examples demonstrating how prompt-engineering techniques can impact LLMs performance on conceptual physics tasks and (3) to discuss potential implications of the understanding of LLMs and prompt engineering for physics teaching and learning. We first summarise existing research on the performance of a popular LLM-based chatbot (ChatGPT) on physics tasks. We then give a basic account of how LLMs work, illustrate essential features of their functioning, and discuss their strengths and limitations. Equipped with this knowledge, we discuss some challenges with generating useful output with ChatGPT-4 in the context of introductory physics, paying special attention to conceptual questions and problems. We then provide a condensed overview of relevant literature on prompt engineering and demonstrate through illustrative examples how selected prompt-engineering techniques can be employed to improve ChatGPT-4's output on conceptual introductory physics problems. Qualitatively studying these examples provides additional insights into ChatGPT's functioning and its utility in physics problem-solving. Finally, we consider how insights from the paper can inform the use of LLMs in the teaching and learning of physics.

La clasificación automática de contenidos de la Dark Web es relevante para la detección e investigación de actividades delictivas. Sin embargo, la escasez de datos etiquetados impone límites al uso de clasificadores supervisados. Los grandes modelos de lenguaje, con capacidad de clasificación en categorías en las que no han sido entrenados, no están sujetos a estas limitaciones. El estudio se centra en evaluar el rendimiento del modelo de lenguaje GPT-3.5 de OpenAI para clasificar contenido de texto de la Dark Web bajo un enfoque de zero-shot prompting. El modelo alcanza un valor F1 ponderado del 80,5%, detectándose grandes diferencias entre categorías y distintas problemáticas que limitan su rendimiento respecto al estado del arte en clasificadores supervisados. El análisis de los errores cometidos permite identificar contextos de aplicación en los que podría resultar competitivo bajo las condiciones de estudio, así como sugerir distintas estrategias para mejorar su rendimiento.

Wondering where generative artificial intelligence (AI) fits in information literacy instruction? This session will share findings from a recent survey of library professionals on how they are already teaching about and using AI powered ChatGPT in information literacy instruction and where they see potential opportunities and areas of concern. Survey analysis will include information about attitudes, current and anticipated use, and descriptions of teaching methods that leverage the technology. As we navigate the survey results, attendees will have the opportunity to share their own perspectives on the same questions via live polling. We will then turn to attendees to share their experiences and the challenges they have found in addressing the needs of their library communities as they explore applications of AI in their research, instruction, writing, and creative endeavors. As we explore the topic we will cover issues ranging from what generative AI is and is not, “hallucinated” information, deep fakes, misinformation, biased training data, documenting use of AI, privacy, copyright and costs.

University students have used ChatGPT extensively to complete their academic tasks since its release in November 2022. The objective of the paper was to explore university students’ perceptions regarding the role of ChatGPT in enhancing their academic performance and their concerns regarding the ethical and social implications of using ChatGPT in their academic tasks. The research design of the current qualitative study was phenomenology. Data were collected using semi-structured interviews with eight participants selected purposefully from a private-sector university in Lahore. We used NVivo software (version 14) for data analysis. Clarke and Braun’s (2013) guide for thematic analysis was followed to read, review, and identify the codes, patterns, and themes. The analysis revealed that the majority of respondents believed that ChatGPT is a very effective tool that offers practical support to them in generating innovative ideas for solving academic problems. Students also showed serious concerns about the lack of clarity and policy guidelines for effectively using ChatGPT in writing their academic tasks. Based on the study findings, we suggest universities should develop and enforce a policy clarifying the limitations and scope of ChatGPT.

Large language models (LLM) are considered by many researchers as promising technology for automating routine tasks. Results from applying LLM in engineering disciplines such as Enterprise Modeling also indicate potential for the support of modeling activities. LLMs are fine-tuned for specific tasks using chat based interaction through the use of prompts. This paper aims at a detailed investigation of the potential of LLMs in Enterprise Modeling (EM) by taking the perspective of EM method adaptation of selected parts of the modeling process within the context of using prompts to interrogate the LLM. The research question addressed is: What adaptations in EM methods have to be made to exploit the potential of prompt based interaction with LLMs? The main contributions are (1) a meta-model for prompt engineering that integrates the concepts of the modeling domain under consideration with the notation of the modeling language applied and the input and output of prompts, (2) an investigation into the general potential of LLM in EM methods and its application in the 4EM method, and (3) implications for enterprise modeling methods.

In many engineering disciplines, modeling is considered an essential part of the development process. Examples are model-based development in software engineering, enterprise engineering in industrial organization, or digital twin engineering in manufacturing. In these engineering disciplines, the application of modeling usually includes different phases such as target setting, requirements elicitation, architecture specification, system design, or test case development. The focus of the work presented in this paper is on the early phases of systems development, specifically on requirements engineering (RE). More specifically, we address the question of whether domain experts can be substituted by artificial intelligence (AI) usage. The aim of our work is to contribute to a more detailed understanding of the limits of large language models (LLMs). In this work, we widen the investigation to include not only processes but also required roles, legal frame conditions, and resources. Furthermore, we aim to develop not only a rough process overview but also a detailed process description. For this purpose, we use a process from hospitality management and compare the output of ChatGPT, one of the most popular LLMs currently, with the view of a domain expert.

Large Language Models (LLMs) are a type of machine learning that handles a wide range of Natural Language Processing (NLP) scenarios. Recently, in December 2022, a company called OpenAI released ChatGPT, a tool that, within a few months, became the most representative example of LLMs, automatically generating unique and coherent text on many topics, summarising and rewriting it, or even translating it to other languages. ChatGPT originated some controversy in academia since students can generate unique text for writing assessments being sometimes extremely difficult to distinguish whether it comes from ChatGPT or a person. In research, some journals specifically banned ChatGPT in scientific papers. However, when used correctly, it becomes a powerful tool to rewrite, for instance, scientific papers and, thus, deliver researchers’ messages in a better way. In this paper, we conduct an empirical study of the impact of ChatGPT in research. We downloaded the abstract of over 45,000 papers from over 300 journals from Dec 2022 and Feb 2023 belonging to different research editorials. We use four of the most known ChatGPT detection tools and conclude that ChatGPT played a role in around 10% of the papers published in every editorial, showing that authors from different fields have rapidly adopted such a tool in their research. © The Author(s) 2024., Funding: Open access funding provided by Halmstad University.

The popularity of Large Language Models (LLMs), such as ChatGPT-4 (GPT), is increasing rapidly. Simultaneously, anxiety is contributing to the global burden of disease. To date, little research is performed on how LLMs portray anxiety. This study aimed to investigate the quality of anxiety portrayals generated by LLMs by “How well do large language models produce realistic anxiety depictions?”, using a mixed-methods approach. Realism was defined as perceived realism and alignment with scientific literature. The quantitative analysis involved 42 participants (age M = 30) randomly assigned to a questionnaire representing one of four anxiety levels. Participants rated perceived level and perceived realism of anxiety descriptions in human-written and GPT-modified texts. Results revealed: GPT-generated anxiety descriptions were perceived significantly more realistic compared to human-written texts; regardless of anxiety level; participants were less accurate in rating anxiety levels of GPTgenerated stories when the levels differed largely from that of the human-written story. The qualitative narrative analysis provided deeper insights into how realistic GPT and a GPT based on the cognitive framework of anxiety (CGPT) depicted anxiety by assessing how well depictions aligned with scientific literature. GPT and CGPT effectively included general features of anxiety in depictions. CGPT focused more on cognitive thought patterns, but neither fully depicted distinctions between different levels of anxiety. Overall, findings suggest LLMs do well in producing realistic representations of anxiety, but fail to fully depict various levels of anxiety. The study contributes to understanding potential applications of LLMs in psychological contexts, including management training, therapeutic settings, and literature. Keywords: LLM, ChatGPT, anxiety, realism, cognitive framework, Språkmodeller (LLMs) som ChatGPT-4 (GPT) blir allt mer populära. Samtidigt påverkar ångest folkhälsan. Hittills har enbart ett fåtal studier utrett hur LLMs skildrar ångest. Syftet med denna studie var att undersöka kvaliteten på ångestskildringar genererade av LLMs med en blandad metodansats genom frågan “Hur väl producerar LLMs realistiska ångestskildringar?”. Realism definierades som upplevd realism och överensstämmelse med vetenskaplig litteratur. I den kvantitativa analysen tilldelades 42 deltagare (ålder M = 30) en enkät som representerade en av fyra ångestnivåer. Deltagarna bedömde upplevd ångestnivå och upplevd realism av ångestbeskrivningar i berättelser skrivna av människor och GPTgenererade berättelser. Resultaten visade att deltagarna uppfattade GPT-genererade ångestbeskrivningar som signifikant mer realistiska, oavsett ångestnivå, men var mindre korrekta i att bedöma ångestnivå i GPT-genererade berättelser när nivån skiljde sig mycket från den människoskrivna versionen. Den kvalitativa narrativa analysen gav djupare insikter i hur realistiskt GPT och en GPT baserad på det kognitiva perspektivet av ångest (CGPT) skildrade ångest genom att jämföra skildringarna med vetenskaplig litteratur. Skildringarna av GPT och CGPT överensstämde i stort sett med litteraturen. CGPT fokuserade mer på kognitiva tankemönster, men ingen modell lyckades fullt ut skildra skillnaderna mellan olika ångestnivåer. Sammantaget tyder resultaten på att LLMs är bra på att skapa realistiska representationer av ångest, men misslyckas med att fullt ut skildra olika ångestnivåer. Studien bidrar till kunskapen om potentiella tillämpningar av LLMs i psykologiska kontexter, såsom ledarskapsutbildning, terapi och inom litteratur. Nyckelord: LLM, ChatGPT, ångest, realism, kognitiv ångest

De AI-drivna språkmodellernas ökade tillgänglighet innebär nya utmaningar och möjligheter inom svenskundervisningen. Med hjälp av verktyg som ChatGPT kan elever generera högkvalitativa texter inom loppet av bara sekunder, vilket har ändrat förutsättningarna för hur uppgifter kan konstrueras i ämnet. Syftet med detta arbete var att utforma och utvärdera en didaktisk design där ChatGPT var den primära resursen för undervisning i berättarteknik och stilistik i svenska. Designen testades genom en intervention där en gymnasieklass fick agera skrivcoacher till ChatGPT och bearbeta noveller med hjälp av litteraturvetenskapliga begrepp. Arbetsmomentet avslutades med muntliga redovisningar där eleverna reflekterade över sina arbeten. En analys av de muntliga redovisningarna visade att ChatGPT verkar ha underlättat för eleverna att göra litteraturvetenskapliga och lingvistiska analyser. Detta främst genom att verktygets interaktiva funktioner gav eleverna agens i arbetet och kunde användas för att visa hur litteraturvetenskapliga begrepp påverkar en skönlitterär text. Undervisningsmetoden medförde även en processorienterad syn på skönlitterärt skrivande och gav insikter om generativ AI som medium. Eleverna upplevde undervisningsmetoden som rolig och betydelsefull, men de ansåg inte att den avsevärt ökade deras arbetsmotivation. Därutöver förekom det att eleverna kopierade tolkningar som hade gjorts av ChatGPT i stället för att utföra egna analyser. Det uppstod även svårigheter med bedömningen då det stundtals var svårt att skilja på elevernas och ChatGPT:s analyser., The increased availability of the AI-driven language models has resulted in new challenges and opportunities for teaching Swedish. With the help of tools like ChatGPT, students can generate high-quality texts in a matter of seconds, which has changed the conditions for how assignments can be constructed in the subject. The purpose of this work was to design and evaluate a didactic design where ChatGPT was the primary resource for teaching storytelling techniques and stylistics in Swedish. The design was tested through an intervention in upper secondary school, where the students acted as writing coaches for ChatGPT and produced short stories using literary concepts. The work phase ended with oral presentations in which the students reflected on their work. An analysis of the oral presentations showed that ChatGPT seems to have made it easier for the students to make literary and linguistic analyses. This is primarily because the tool's interactive functions gave the students agency in the work and could be used to show how literary concepts affect a fictional text. The teaching method gave the students a process-oriented view on writing fiction and provided insights into generative AI as a medium. The students viewed the teaching method as fun and meaningful, but they did not feel that it significantly increased their work motivation. In addition, some students copied interpretations made by ChatGPT instead of performing their own analyses. There were also difficulties with the assessment as it was sometimes difficult to distinguish between the students' and ChatGPT's analyses.

AI-assistenter har blivit mycket populära bland konsumenter, och organisationer inser i allt högre grad deras potentiella fördelar. Experter menar att implementering av AI-assistenter stöter på samma hinder som vid implementering av vanliga IT-system, men att AI-assistenter har unika aspekter som försvårar implementering ytterligare. Tidigare forskning har huvudsakligen baserats på expertåsikter och det saknas omfattande studier om specifika hinder som behöver åtgärdas för en framgångsrik implementering. Denna studie syftar till att fylla denna kunskapslucka genom att intervjua tolv deltagare från ett IT-företag för att få en helhetsförståelse av deras perspektiv på AI-assistenter i deras yrkesmiljö. Med Unified Theory of Acceptance and Use of Technology 2 (UTAUT2) som teoretiskt ramverk, analyserades kvalitativa data från semistrukturerade intervjuer. Resultaten visar att deltagarna generellt har en positiv inställning till AI-assistenter, men de uttrycker flera utmaningar som kan försvåra implementeringen av dem. Dessa utmaningar inkluderar svårigheter att identifiera lämpliga tillämpningsområden, skiftande uppfattning av användningsområden, obekantskap med teknologin, stress relaterad till implementering, säkerhets- och tillitsfrågor, samt begränsade resurser., AI assistants have become very popular among consumers, and organizations are increasingly recognizing their potential benefits. Experts suggest that implementing AI assistants faces the same obstacles as implementing regular IT systems, but with unique aspects that further complicate implementation. Previous research has mainly been based on expert opinions, and there is a lack of comprehensive studies on specific obstacles that need to be addressed for successful implementation. This study aims to fill this knowledge gap by interviewing twelve participants from an IT company to gain a comprehensive understanding of their perspectives on AI assistants in their professional environment. Using the Unified Theory of Acceptance and Use of Technology 2 (UTAUT2) as a theoretical framework, qualitative data from semi-structured interviews were analyzed. The results show that participants generally have a positive attitude towards AI assistants but express several challenges that could hinder the implementation. These challenges include difficulties in identifying appropriate application areas, shifting perceptions of use cases, unfamiliarity with the technology, stress related to implementation, security and trust issues, and limited resources.

Den snabba utvecklingen av Artificiell Intelligens (AI) har lett till att organisationer och företag i en högre grad väljer att integrera AI-drivna lösningar i sin verksamhet i syfte att förbli konkurrenskraftiga och öka produktiviteten. Framsteg inom AI har haft en betydande inverkan på HR-branschen vilken tidigare forskning har visat att AI har potential att effektivisera särskilda HR-processer. Mot denna bakgrund var syftet med studien att undersöka HR-anställdas upplevelser av att använda det AI-drivna verktyget ChatGPT i sitt arbete. Studien genomfördes genom kvalitativ datainsamling med semistrukturerade intervjuer. Resultatet analyserades genom en tematisk analys som resulterade i fyra teman: effektivitet och tidsbesparing, kvalitet, ChatGPT som interaktiv resurs och oro för AI-utvecklingen. Både positiva och negativa upplevelser av att använda ChatGPT inom HR framkom i resultatet, inklusive AI:s potential att effektivisera och spara tid, samt riskerna med verktygets partiskhet, integritetsskydd och eventuell förlust av den mänskliga aspekten i HR. Studien bidrar till en bredare förståelse för AI:s roll i HR med de potentiella möjligheter och utmaningar som följer med dess implementering.

Studien behandlar respons på elevtexter där målet är att bidra till en diskussion om AI som hjälpmedel på gymnasiet. Därför är syftet med studien att explorativt utforska potentialen hos AI och dess genererade respons på elevers utredande texter i svenska 3 jämfört med kvaliteten hos respons given av lärare. Materialet består av tolv elevtexter, två från varje betygsnivå (A-F), insamlade från fyra lärare med varierande antal texter per lärare. Analysmetoden kombinerar kvalitativ textanalys av lärarresponsen med en explorativ jämförande analys av AI-responsen. Det teoretiska ramverket består av en kombination av Hattie och Timperleys (2007) responsmodell samt Cerratto Pargmans och Jahnkes (2019) begrepp om aktörer. Resultatet visar att lärarresponsen på elevtexterna varierar i stadium och nivå, med Feed Up som den vanligaste responsen och Feed Back som den minst förekommande. ChatGPT tenderar att inte ge liknande bedömningar som lärare och saknar ofta den detaljerade och specifika återkoppling som lärarna ger, särskilt när det gäller att främja elevens utveckling och nyansering av resonemang. Studien kommer fram till att det finns stor potential för generativ AI i skolan, men med vissa begränsningar i nuläget., The study addresses feedback on student texts, aiming to contribute to the discussion of AI as an aid in upper secondary school education. Consequently, the purpose of the study is to exploratively investigate the potential of AI and its generated responses to students' analytical texts in Swedish 3, compared to the quality of feedback provided by teachers. The material consists of twelve student texts, two from each grade level (A-F), collected from four teachers with varying numbers of texts per teacher. The analysis method combines qualitative text analysis of the teacher feedback with an exploratory comparative analysis of the AI feedback. The theoretical framework consists of a combination of Hattie and Timperley's (2007) feedback model and Cerratto Pargman and Jahnke's (2019) concept of agency. The results show that the teacher feedback on student texts varies in stage and level, with Feed Up being the most common response and Feed Back being the least frequent. ChatGPT tends not to provide similar assessments as teachers and often lacks the detailed and specific feedback that teachers give, especially when it comes to promoting the student's development and nuanced reasoning. The study concludes that there is great potential for generative AI in schools, but with certain limitations at present.

This study examines the efficacy of using artificial intelligence (AI) generated questions andfeedback to enhance learning through retrieval practice (RP) compared to traditionalrestudying (S). Additionally, the study explored the correlation between students' attitudestowards AI and learning outcomes as measured immediately after learning and two days afterthe learning intervention. Conducted with 18 participants from Umeå University, the researchevaluates the potential of ChatGPT in facilitating RP and improving retention andcomprehension of text-based materials. Participants were presented with two texts and eachtext was assigned to either the RP or S condition. In the RP condition, participants answeredAI-generated questions with feedback, in the S condition, participants reread the texts. Resultsindicate that RP significantly outperformed S in the day 1 and day 3 tests, aligning withestablished research on the testing effect (TE). The authors suggest that AI, as ChatGPT, caneffectively support RP by generating questions and providing dynamic and personalizedfeedback, potentially reducing educators' workload. It highlights the importance of furtherresearch to optimize AI-generated content and feedback mechanisms and to explorelonger-term memory effects. The findings advocate for integrating AI tools in educationalsettings to enhance active learning and retention, and stresses the need for ongoingexploration of AI's role in education.

Titel: Kan artificiell intelligens skapa en trovärdig hållbarhetsrapport? En kvantitativ studie om förmågan att urskilja AI-genererade texter från mänskligt skrivna texter samt vilka faktorer som påverkar. Nivå: Examensarbete på grundnivå (kandidatexamen) i ämnet företagsekonomi Författare: Arvid Högberg och William Karlström Handledare: Asif M. Huq Datum: 2024 – maj Syfte: Experimentet syftar till att se om faktorer som yrkesmässig bakgrund, utbildningsnivå, ålder och kön påverkar respondenternas förmåga att urskilja AI-genererad text från mänskligt framställd text. Metod: Studien utfördes med en kvantitativ enkätundersökning där 84 respondenter deltog. Den bestod av inledningsvis demografiska frågor som följdes av två experiment där två texter presenterades och respondenten avgjorde vilken text som var AI-genererad. Analysmetoderna som använts var ANOVA (Analysis of Variance), frekvenstabeller och korrelationsanalyser. Våra respondenter var ekonomistudenter, utbildare inom ekonomi och yrkesverksamma inom redovisning/revision. Resultat och slutsats: Studien visade att faktorer som kön, ålder och erfarenhet inte hade någon signifikant påverkan på förmågan att urskilja AI-genererad text. Men högre AI-användning ökade chansen att urskilja AI-genererad text. Examensarbetets bidrag: Studien har bidragit med värdefull information om vad som påverkar förmågan att urskilja AI-genererad text från mänskligt skriven text. Resultatet visar att demografiska faktorer inte spelar någon signifikant roll. AI-användning däremot har en viss påverkan. Studien visar även hur långt AI-teknologin har kommit och hur svårt det kan vara att skilja mellan AI-genererad text och mänskligt skriven text. Förslag till fortsatt forskning: Effekterna av AI-träning och utbildning av människors förmåga att skilja på olika typer av text är något som framtida forskning behöver belysa. Det går även att utforska användningen av AI för textgenerering inom olika branscher och hur detta påverkar människo, Title: Can artificial intelligence create a credible sustainability report? A quantitative study on the ability to distinguish AI-generated texts from human-written texts and which factors influence. Level: Bachelor's degree thesis in business administration Authors: Arvid Högberg and William Karlström Supervisor: Asif M. Huq Date: 2024 – May Purpose: The experiment aims to see if factors such as work experience, education background, age and gender affect the respondents' ability to distinguish AI-generated text from human-produced text. Method: The study was carried out with a quantitative survey in which 84 respondents participated. It consisted of initial demographic questions followed by two experiments where two texts were presented and the respondent decided which text was AI-generated. The analysis methods used were ANOVA (Analysis of Variance), frequency tables and correlation analyses. Our respondents were economics students, educators in economics and professionals in accounting/auditing. Results and conclusion: The study showed that factors such as gender, age and experience had no significant impact on the ability to distinguish AI-generated text. But higher AI usage increased the chance of distinguishing AI-generated text. Contribution of the thesis: The study has contributed valuable information about what affects the ability to distinguish AI-generated text from human-written text. The result shows that demographic factors do not play a significant role. AI use, on the other hand, has a certain impact. The study also shows how far AI technology has come and how difficult it can be to distinguish between AI-generated text and human-written text. Suggestions for further research: The effects of AI training and education on people's ability to distinguish between different types of text is something that future research needs to explore. It is also possible to explore the use of AI for text generation in various industries and how this affects people's

Den föreliggande studie syftade till att studera universitetsstudenters åsikter om användningen av ChatGPT i studier. Studien fokuserade på tre frågeställningar: i vilken omfattning studenter använder ChatGPT, deras moraliska bedömning av användningen samt eventuella skillnader mellan studenter. En kvantitativ forskningsansats användes i studien och datamaterialet baserades på en enkät som delades ut via digitala plattformar till universitetsstudenter. Enkätsvaren analyserades sedan med hjälp av beskrivande statistik analys för att identifiera och förstå samband mellan olika variabler. Resultatet visade att en majoritet av studenterna, 81%, använder ChatGPT i sina studier. De flesta studenter anser att ChatGPT är ett värdefullt komplement som hjälper dem att förstå komplexa problem, sammanfatta information och spara tid. Trots dessa fördelar framhöll många studenter vikten av ett ansvarsfullt användande för att undvika risker som fusk och minskat lärande. En betydande del av studenterna upplevde en osäkerhet kring de moraliska aspekterna av att använda ChatGPT, men valde ändå att använda verktyget på grund av dess praktiska fördelar. Studien fann inga statistiskt signifikanta skillnader i användningen av ChatGPT baserat på ålder, kön eller föräldrarnas utbildningsnivå. Däremot var studenter från tekniska och samhällsvetenskapliga program något mer benägna att använda AI-verktyget, vilket med stöd av Bourdieus teori om kapital och habitus kan indikera att studenter med högre kulturellt kapital och vana vid tekniska hjälpmedel är mer öppna för nya innovationer. Det fanns även en statistiskt signifikant skillnad i användningen av ChatGPT baserat på studenternas moraliska uppfattning om AI-verktyget, vilket visar på en koppling mellan moraliska bedömningar och användningen av ChatGPT. Slutligen betonar studien vikten av att utveckla studenternas kritiska tänkande och att använda ChatGPT på ett ansvarsfullt sätt för att inte negativt påverka inlärningen. Lärarnas inställ

Following the invention of Artificial Intelligence (AI) and its rapid dissemination across the globe, organisations have been experiencing a significant shift in processes, performance, and business models. As a result of the hype created by the release of ChatGPT, both the business and academic world has approached AI with a new and enormous interest. However, studies about different aspects in relation to perceptions and attitudes towards AI adoption in the private sector, and particularly, case studies are limited. Aim: This Bachelor’s thesis aims to analyse the perceptions of management, IT management, and frontline employees about the implementation and impact of AI in large companies, focusing on Scania as a representative of a large automotive multinational corporation. Through the study, we attempt to explore the perceptions about a range of aspects about AI, such as its impact on change in organisation, impact on job and skills, trust in AI, AI opportunities and risk. We also aim to understand the effect of the advent of ChatGPT on perceptions about AI and its integration in organisation. Methodology: This thesis is a qualitative case study that employed semi-structured interviews for collection of empirical data. The sample was generated through convenience sampling and included representatives of the company of following groups: IT-managers, managers, and non-managerial employees. This ensured a broad representation of views concerning AI implementation and impact in Scania. Conclusion and Contribution: Our findings reveal that AI implementation and impact perceptions range depending on job roles, primarily when it comes to the perceptions towards AI usefulness, risks and opportunities, and its impact on jobs. Additionally, we argue that the launch of ChatGPT has enhanced AI’s awareness and acceptance as a new technology. The thesis can be utilised as a basis for future studies on perceptions and factors affecting AI implementation and impact of represe

Focusing on personalized grammar tasks, this study dives into the integration of Generative Artificial Intelligence into English as a Second Language education. By utilizing a mixed methods approach, incorporating both qualitative and quantitative analyses the study explores how personalized learning can be improved by employing ChatGPT. Results from the study indicate that GAI-driven personalization significantly enhances student engagement and motivation. This offers a promising path for tailoring education to individual learner needs toward a more inclusive classroom. A central outcome of this study is the proposal of a new theoretical framework the Personalization-Motivation Integration Framework (PMIF). This framework clarifies the synergistic effects of integrating content and topic personalization to significantly boost student motivation and reach a more inclusive learning environment. This adds to the growing research about AI's potential in education as it indicates that these technologies can significantly enhance teaching and offer a more tailored and inclusive learning environment.

Syftet med studien är att bidra med ökad kunskap om gymnasieelevers erfarenheter och upplevelser av den digitala teknikutvecklingen, inklusive artificiell intelligens och algoritmiska tjänster, och hur elever formerar sina tankar om vilka konsekvenser det får för samhällskunskapsundervisningen och det framtida demokratiska deltagandet. För att uppfylla syftet har semistrukturerade samtalsintervjuer använts som datainsamlingsmetod där totalt elva elever intervjuats från två olika skolor. En öppen tematisk analys applicerades på datan som resulterade i åtta olika teman. Därefter analyserades resultatet utifrån Andrew Feenbergs teoretiska referensram som består av en kritisk förståelse för hur teknologi och dess utveckling formar och påverkar samhället och hur människor interagerar med den. Eleverna kom fram till att AI kan användas för att effektivisera industrin likväl som ett verktyg i skolan för att hjälpa elever med inlärningen. Däremot har AI negativa sidor också. Rädslan att AI tar över mänskligheten växer och det skapas innehåll som är deepfake där det är svårt att avgöra vad som stämmer. Algoritmer kan också bli ett problem då det många gånger baseras på vad du tittar på och att det blir som en konfirmeringsbias. Du får bara upp ett fåtal perspektiv. Elevernamenar att människor utnyttjas av producenterna av AI och algoritmer eftersom det styr oss i en viss riktning. Eleverna kommer fram till att det måste finnas någon kontroll för att inte kränka individer och skada demokratin och ger förslag på en statlig eller överstatlig kontroll. AI har även lett till att eleverna får en försvårad skolgång med fler prov. Tilliten mellan lärare ochelever minskar. AI måste integreras mer i skolan. Därtill menar eleverna att nya kunskaper kommer bli viktiga, däribland tekniska kunskaper, källkritik, öppenhet och förmågor så som att resonera och argumentera. Eleverna har en önskan om att de ska förberedas med de nya kunskaperna för att aktivt kunna ta del av det nya samhället, The purpose of the study is to contribute with knowledge about high school students’ experiences of the development of digital technology, including artificial intelligence and algorithms. The study aims to explain how the development of digital technologies might change the subject social studies in school and a future democratic citizenship. Semi-structuredconversational interviews have been used as a data collection method where a total of eleven students have been interviewed from two different schools. An open thematic analysis was used which resulted in eight themes. The results were analyzed based on Andrew Feenberg’s theory, which consist of a critical understanding of how technology and its development shape and influence society and how people interact with it. The students concluded that AI can be used as a tool to help students with learning. However, students felt a rising fear that AI might wipe out humanity and the fact that it can create deepfake content which makes it difficult to determine what is true. Algorithms, like your for-u-page can lead to a narrowed perspectivewhere you do not see the bigger picture. Furthermore, the students believe that humans are exploited by the producers of AI and platforms that uses algorithms since it steers us in a certain direction. The students concluded that AI needs to be controlled, for example by the state. They think that might avoid violations and a restriction of democracy. The development of AI hasled to an increase in tests for the students. The trust between teachers and students have decreased. The students wish to talk more about AI and the effects of it. They also believe that a new set of competences will become more important, for example technical knowledge, how to deal with a big amount of information as well as abilities such as reasoning and arguing, to be able to actively take part in the new society that is yet to come.

Identifying the most used design patterns and communication methods for a microservices architecture holds paramount importance in ensuring system stability within the IT industry. With a plethora of approaches available, encompassing various design patterns and communication techniques, the selection of technologies often hinges upon the expertise and knowledge of software designers. However, designing microservices solely through human efforts poses significant challenges for companies. Moreover, the monitoring aspect presents its own set of complexities in the realm of microservices. This research aims to elucidate the prevalent design patterns and communication methods utilized in modern IT microservices environments. It also seeks to demonstrate the integration of Artificial Intelligence tools such as ChatGPT into the architecture design process, particularly in enhancing monitoring capabilities within microservices. Subsequently, a prototype is proposed to offer design guidance, as recommended by the ChatGPT and Llama 2 customized models in accordance with the specified requirements. Additionally, the study primarily seeks to identify prevalent design patterns and communication methods employed within the IT industry through the utilization of surveys and interviews conducted with both technical and non-technical personnel.

This study examines both the opportunities and challenges of including ChatGPT in education from the perspective of middle and high school teachers. The main objective is to highlight not only the challenges posed by AI but also the potential benefits of AI in education, as negative opinions about ChatGPT are more commonly heard. The study applies the epistemological perspectives of sociocultural and pragmatism to gain a deeper understanding of teachers’ views on the opportunities and challenges that ChatGPT can bring to education. The results show that ChatGPT offers several opportunities as well as challenges, which are explained in more detail in the study. The most interesting finding is that critical thinking emerges as a common factor in the teachers’ opinions, the ability to critically evaluate information has become more important with the introduction of ChatGPT. Teachers believe that ChatGPT can serve as a valuable complement in education, provided that students learn to use the tool in a critical and responsible manner. This can enhance students’ critical thinking and analytical skills, which are crucial for their development.

De senaste åren har generativa AI-modeller såsom ChatGPT blivit ett populärt fenomen inom fleraområden, inklusive utbildning inom programmering. Frågor som uppstår i samband med detta är hur verktyg som ChatGPT kan påverka studenter samt hur utbildning kan anpassas med tanke på verktyget inverkan. Denna studie har valt att rikta in på hur användning av generativ AI påverkar studenters inlärningsmotivation och tilltro till att klara av programmeringsuppgifter. Studien syftar även till att fånga in studenters perspektiv kring hur de uppfattar att man bör anpassa utbildning för att hantera detta nya fenomen med utgångspunkt i utlärning av prompter. Studien har kommit fram till att studenterna som intervjuats har upplevt ökat tilltro till att klara av och lösa uppgifter med vid användning av generativ AI. Flera av studenterna hade också upplevt en ökning i inlärningsmotivationen på grund av generativ AI. Perspektiven studenterna hade gällande anpassning av utbildning uppfattade flera att utlärning av prompter som ett effektivt sätt att integrera generativ AI inom programmeringsutbildningar. Samtidigt uppfattade andra studenter vissa aspekter som viktigare att fokusera på gällande hur utbildning bör förhålla sig till generativ AI där bland annat ökad redovisning av uppgifter och ändring av svårighetsgraden på uppgifter beskrevs som viktigare anpassningar.

Framväxten av artificiell intelligens (AI) och generativ AI har lett till lanseringen av AI-baserade chatbottar som OpenAI:s ChatGPT. Det är i stort sett fortfarande oklart vilken effekt användningen av dessa verktyg har på undervisning i engelska som främmande språk. Den empiriska undersökningen som presenteras i detta examensarbete har därför syftet att undersöka vilken effekt ChatGPT-feedback har i jämförelse med lärarfeedback på svenska gymnasieelevers komplexitet, korrekthet och flyt i skriftlig produktion på engelska. Ett ytterligare mål med denna studie var att undersöka om elevernas språkliga kompetens påverkar dessa effekter. Studien genomfördes genom att undersöka 56 gymnasielevers argumenterande texter före och efter feedback. Studiens resultat pekar på att båda feedbackmodaliteter har liknande effekt på elevernas komplexitet, korrekthet och flyt. Statistiskt signifikanta ökningar går att se i dimensionerna för korrekthet, lexikal komplexitet och flyt för båda feedbackmodliteter. För dimensionen syntaktisk komplexitet har endast ChatGPT gruppen nått en statistiskt signifikant ökning, och för korrekthet har Lärarfeedback haft en större effekt med statistiskt signifikanta ökningar inom alla fyra analytiska mått för dimensionen. Vidare visar studien att elever med lägre språklig kompetens gynnas mer av feedbacken än elever med högre språkliga kompetens. Studien belyser även behovet av att framtida forskning fokuserar på ChatGPT-feedbacks effekter på texters generella kommunikativa funktion för att bättre förstå vilka effekter AI-genererad feedback har på skriftlig produktion i engelska som främmande språk.

This study aims to investigate how communication specialists perceive the integration of generative artificial intelligence (AI) within their profession and how these changes affect their professional role. The purpose of the study was achieved by conducting qualitative interviews with eight communication professionals regarding their view on the integration of generative AI within their profession. Particularly, we found it interesting to examine how the communication specialists use generative AI within their professional role, what risks and/or opportunities they identify concerning the use of generative AI tools within the communication profession and what factors they deem as crucial for successfully and effectively integrating AI tools within their professional roles. The theoretical framework of the study consists of diffusion theory, actor-network theory, risk society and reflexive modernity. The results of the study indicate that communication specialists see several opportunities in integrating generative AI tools into their work. Key opportunities identified included using generative AI for inspiration and automating repetitive tasks, thereby allowing communication professionals to allocate more time to the strategic aspects of their work. There was no expressed concern that generative AI could currently replace the communication specialist’s professional role. Thus, the risks identified did not seem to revolve around the job security of the communication specialists but rather around the uncertainty about how generative AI functions, its capabilities, and how it handles the input data. Therefore, the establishment of clearer guidelines and policies on AI usage was something that the communication specialists in this study expressed a great need for. Additional factors identified by communication specialists as crucial for successful AI integration included individual characteristics such as curiosity, a willingness to learn, and openness to change. Thus

Our study investigates the possibility of integrating ChatGPT, an AI text generative tool, as a didactic supporting tool within the Swedish upper-secondary school curricula. Particularly, it takes interest in the English foreign language teaching context and examines eight peer reviewed articles on the matter. The questions it aims to answer are: In light of existing research, 1) what are some of the implications of integrating ChatGPT as a didactic supporting tool in Sweden’s English as a Foreign Language (EFL) upper-secondary classroom and 2) how could this tool be incorporated successfully into the teaching process? In the results, the synthesis identifies ethical concerns, teacher beliefs and perspectives, and pros and cons of integration as the three primary themes and shows how 75% of the articles are concerned with the ethical implications of ChatGPT, hence requesting ethical frameworks and guidelines. Additionally, it identifies teachers' beliefs and attitudes on integration of AI as the biggest influencing factor on the integration of ChatGPT and that the dominant benefits of integration of ChatGPT are in student accessibility and reduced workload for teachers. Conversely, the concerns include those of unethical nature such as cheating, plagiarizing, and information security and accuracy. The discussion suggests that ChatGPT, as a didactic tool, is supported by the Swedish upper-secondary curricula. However, additional efforts are needed to ensure its safe and effective integration. The study concludes that ChatGPT, and other AI generative tools, could be implemented in the curricula, but before this is accomplished we require policies, guidelines, and ethical frameworks to ensure it can be done safely and effectively.

This ethnographic study investigates how ChatGPT meets information needs, andthe difference in functional outcomes when using ChatGPT in a reference transaction compared to a librarian. The theoretical starting point is Taylor's theories on information needs and filters, commands and questions. The study is based on empirical material consisting of observations of libraries and of ChatGPT. The empirical material is analyzed using analytical autoethnography, thick description, and Childer's system for evaluating reference transactions. The purpose of the thesis is to investigate how ChatGPT performs in reference transactions compared to librarians. The study shows that ChatGPT meets information needs in a different way than librarians. The biggest difference in functional results is that ChatGPT's responses are incorrect more frequently than the librarians. ChatGPT not only makes dubious claims, it also hallucinates and recommends books that do not exist. This thesis can be used as a basis for further research, as well as support for public libraries that want to use AI technology in reference transactions.

Artifical intelligence is a part of the progress of the digitalization in the society. Language models such as ChatGPT have the ability to answer questions posed by the user as well as to produce texts that resemble human writing. However, the use of language models is not without risk. They can be used as a tool to encourage cheating on school assignments or to contribute to disinformation. A consequence of the development of AI is that knowledge is obtained in different ways compared to the past, which has to come to transform the traditional teaching methods in schools. The purpose of the study is to investigate how AI affects the tuition in social studies and also what experience social studies teachers have of AI in connection with teaching. The study focuses in upper secondary school in Sweden. The empirical material has been collected through semi-structured interviews. The results of the study show that the teachers see both opportunities and challenges with AI in teaching. AI can be used as a resource by both teachers and students according to the teachers participating in this study. However the result shows that AI also can have an negative impact on the students learning process. Regardless, teachers have been forced to make changes in their teaching.

In the Fourth Industrial Revolution, journalism cannot remain oblivious to the innovations brought about by the socio-technological reality. Artificial intelligence (AI) is one of the technologies on the rise, although the Spanish media have incorporated it little, experimentally and in specific sections. This research has two objectives: 1) to know the state of development of AI in the Spanish digital media and the industry's perception of the impact of this technology on the profession; 2) to explore a human vs synthetic text generation detection method to assess whether generative AI is being used in media. A comparative study was carried out between two media of reference in Spain, one matrix and conservative ideology (El Mundo), and another digital native and progressive trend (eldiario.es). The methodology was based on methodological triangulation and included a survey and an experiment with ChatGPT to detect whether a sample of texts published in these two media was created by humans or machines. The results diagnose that none of these newspapers apply AI, although they are designing plans to do so, and that there seems to be a lack of AI experts in the newsrooms, although a significantly higher predisposition towards AI in the native media is not confirmed as opposed to the matrix. The analyses carried out confirm that generative AI is not used to create texts. The method applied can be used in other studies because its validity is confirmed to provide clues about the human or artificial origin of journalistic texts., En la Cuarta Revolución Industrial el periodismo no puede permanecer ajeno a las innovaciones que configuran una nueva realidad sociotecnológica. La inteligencia artificial (IA) es una de las tecnologías en auge, aunque los medios españoles la han incorporado poco, de forma experimental y en secciones concretas. Esta investigación tiene dos objetivos: 1) conocer el estado de desarrollo de la IA en la prensa digital española y la percepción de la industria sobre el impacto de esta tecnología en la profesión; 2) explorar un método de detección de generación de textos humanos vs sintéticos para evaluar si se utiliza IA generativa en los medios. Se realizó un estudio comparativo entre dos medios de referencia en España, uno matricial y de ideología conservadora (El Mundo), y otro nativo digital y de tendencia progresista (eldiario.es). La metodología se basó en la triangulación metodológica y contempló una encuesta y un experimento con ChatGPT para detectar si textos publicados en estos dos medios fueron creados por humanos o máquinas. Los resultados diagnostican que ninguno de estos periódicos aplica IA, si bien están diseñando planes para hacerlo, y parece haber una carencia de expertos en IA en las redacciones, aunque tampoco se confirma una predisposición significativamente mayor hacia la IA en el medio nativo frente al matricial. Los análisis realizados confirman que no se utiliza IA generativa para crear textos. El método utilizado puede ser empleado en otros estudios porque se confirma su validez para aportar indicios sobre el origen humano o artificial de textos periodísticos.

The aim of this article is to analyze how the Spanish press treats artificial intelligence (AI) and, specifically, the latest generation chatbots, following the launch of the free version of ChatGPT in December 2022. In order to determine this, authors sample the eight national newspapers with the highest sales or number of users, both hybrid (El País, La Vanguardia, ABC, El Mundo) and digital natives (El Español, Okdiario, elDiario.es, Público). From each of these sources, the latest 20 news articles related to AI (n=160) are selected and subjected to a rigorous content analysis using a predefined framework. In addition, a sentiment analysis is performed by ChatGPT and BingChat to identify the predominant attitude of the press towards the AI (positive, negative, or neutral). The findings allow us to conclude, among other results, that the treatment is eminently informative and rarely interpretive. Moreover, the texts primarily focus on the possible uses of this technology; they mainly resort to sources from economic, pedagogical and scientific fields; and they adopt a basically neutral attitude towards AI., El artículo analiza el tratamiento que la prensa generalista española dispensa a la inteligencia artificial (IA) y, de manera específica, a los chatbots de última generación, a raíz del lanzamiento de la versión gratuita de ChatGPT en diciembre de 2022. Para determinarlo, se toman como muestra los ocho diarios nacionales con mayor difusión o número de usuarios, tanto híbridos (El País, La Vanguardia, ABC, El Mundo) como nativos digitales (El Español, Okdiario, elDiario.es, Público), y de cada uno de ellos se seleccionan las últimas 20 noticias publicadas sobre IA (n=160), a las que se aplica una ficha de análisis de contenido. Además, se somete a los textos a un análisis de sentimientos ejecutado por ChatGPT y BingChat, para identificar su actitud —positiva, negativa o neutra— ante el fenómeno. Los resultados permiten concluir, entre otras constataciones, que el tratamiento es eminentemente informativo y rara vez interpretativo; que los textos atienden en especial a las posibles aplicaciones de esta tecnología; que recurren sobre todo a fuentes procedentes de los ámbitos económico, pedagógico y científico; y que adoptan una actitud neutra ante la IA.

Generative artificial intelligence offers opportunities to optimize activities and innovate in the operational tasks of an organization. In the educational field it can be used to enhance the teaching and learning process through its efficient use. The aim of this study was to compare the quality and accuracy of the content generated by generative artificial intelligence tools with the scientific arguments published by researchers to understand the critical and interpretative capacity of the tools. Three text-generating tools were explored using a self-reflection methodology: ChatGPT, Bard and Bing Chat to reflect on the quality and efficiency of teaching-oriented content in the field of training in administration in three blocks: 1) Training in administration, 2) Pedagogical strategies, 3) Research activities. All three tools showed their ability to emulate human responses in surprising ways, however, the information generated lacks technical details and insights to support a more solid content in the responses. It was also identified that the arguments generated by the tools are of skillful use in guiding the use of management methodologies in the educational didactics in administration, so it was deduced that the combination of these perspectives and the support of the scientific literature strengthens the theoretical and practical foundations of the information. This intelligence has the potential to become a valuable tool for transmitting and understanding information, but should not replace the creativity, intellectual and critical work of academic actors., La inteligencia artificial generativa ofrece posibilidades de optimizar actividades e innovar en las tareas operativas de una organización. En el campo educativo puede emplearse para potenciar el proceso de enseñanza y aprendizaje a través de su uso eficiente. El objetivo de este estudio fue comparar la calidad y precisión del contenido generado por herramientas de inteligencia artificial generativa con los argumentos científicos publicados por investigadores para comprender la capacidad crítica e interpretativa de las herramientas. Mediante una metodología de autorreflexión se exploraron tres herramientas generadoras de textos: ChatGPT, Bard y Bing Chat a fin de reflexionar sobre la calidad y eficiencia de contenidos orientados al ejercicio docente en el campo formativo en administración en tres bloques: 1) Formación en administración, 2) Estrategias pedagógicas, 3) Actividades de investigación. Las tres herramientas mostraron su capacidad para emular respuestas humanas de forma sorprendente, sin embargo, la información generada carece de detalles y perspectivas técnicas que respalden un contenido más sólido en las respuestas. También se identificó que los argumentos generados por las herramientas son de gran utilidad para orientar el empleo de metodologías de gestión en las didácticas formativas en administración, por lo que se dedujo que la combinación de esas perspectivas y el respaldo de la literatura científica fortalece la fundamentación teórica y práctica de la información. Esta inteligencia tiene el potencial de convertirse en una valiosa herramienta para transmitir y comprender la información, pero no debe reemplazar la creatividad, el trabajo intelectual y crítico de los actores académicos.

ChatGPT has become a popular technology among people and gained a considerable user base, because of its power to effectively generate responses to users requests. However, as ChatGPT’s popularity has grown and as other natural language processing systems (NLPs) are being developed and adopted, several concerns have been raised about the technology that could have implications on user trust. Because trust plays a central role in user willingness to adopt artificial intelligence (AI) systems and there is no consensus in research on what facilitates trust, it is important to conduct more research to identify the factors that affect user trust in artificial intelligence systems, especially modern technologies such as NLPs. The aim of the study was therefore to identify the factors that affect user trust in NLPs. The findings from the literature within trust and artificial intelligence indicated that there may exist a relationship between trust and transparency, explainability, accuracy, reliability, automation, augmentation, anthropomorphism and data privacy. These factors were quantitatively studied together in order to uncover what affects user trust in NLPs. The result from the study indicated that transparency, accuracy, reliability, automation, augmentation, anthropomorphism and data privacy all have a positive impact on user trust in NLPs, which both supported and opposed previous findings from literature.

Background: Unit testing is essential in software development, ensuring the functionality of individual components like functions and classes. However, manual creation of unit test cases is time-consuming and tedious, impacting testing efficiency and reliability. Problem: Automated unit test generation tools such as EvoSuite and Randoop have addressed some challenges, but they’re limited by language specificity and predefined algorithms. Generative AI tools like ChatGPT and GitHub Copilot powered by OpenAI’sGPT-3.5/4 offer alternatives, but face limitations like user input reliance and operational inconveniences. Solution: CodiumAI’s Codiumate IDE plugin aims to mitigate these limitations, making code quality assurance easier for developers. This study evaluates Codiumate’s trustworthiness in generating unit tests for the Python functions. Method: We randomly selected thirty functions from OpenAI’s HumanEval dataset, and wrote selection criteria for relevant test cases based on each function’s doc string to evaluate Codiumate’s trustworthiness using metrics such as Relevance Score, false positive rate, and result consistency rate. Result: Among all the suggested test cases by Codiumate, 208 unit tests, which consists of 48% of suggested test cases that were relevant. 70% of assertions from these test cases strictly meet selection criteria, while the other 30% while relevant were selected due to our basis and experience in software testing. The average false positive rate is15%. Function groups that have higher Relevance Scores are non-mathematical nature, and simple dependencies. High false positives arise in functions with string and float parameters. All generated unit tests are syntax-error-free, with 20% fail and 80% passed in all five test execution. Conclusion: Codiumate demonstrates potential in automating unit test generation, offering a convenient means to support developers. However, it is not yet fully reliable for critical applications without developer

This study investigates the potential of Large Language Models (LLMs) to automate and enhance software maintenance tasks, focusing on bug detection and code refactoring. Traditional software maintenance, which includes debugging and code optimization, is time-consuming and prone to human error. With advancements in artificial intelligence, LLMs like ChatGPT and Copilot offer promising capabilities for automating these tasks. Through a series of quasi-experiments, we evaluate the effectiveness of ChatGPT 3.5, ChatGPT 4 (Grimoire GPT), and GitHub Copilot. Each model was tested on various code snippets to measure their ability to identify and correct bugs and refactor code while maintaining its original functionality. The results indicatethat ChatGPT 4 (Grimoire GPT) outperforms the other models, demonstrating superior accuracy and effectiveness, with success percentages of 87.5% and 75% in bug detection and code refactoring respectively. This research highlights the potential of advanced LLMs to significantly reduce the time and cost associated with software maintenance, though human oversight is still necessary to ensure code integrity. The findings contribute to the understanding of LLM capabilities in real-world software engineering tasks and pave the way for more intelligent and efficient software maintenance practices.