ABSTRACT. The rapid diffusion of Generative Artificial Intelligence (GenAI) is reshaping the conditions of learning, not by introducing new activities but by reorganising existing feedback structures among students, teachers, and learning environments. This paper adopts a cybernetic perspective to explore how GenAI influences recursive processes of inquiry, collaboration, and reflection. Drawing on concepts of feedback, adaptation, and regulation, the paper frames GenAI as an active participant in learning systems, redistributing agency and reshaping the rhythm of interaction. A methodological guide is proposed for constructing case studies across disciplines, combining documentary artefacts and participant accounts to map cybernetic dynamics in practice. A case study of a Quality Management module illustrates how GenAI reconfigures individual reflection, group collaboration, and assessment practices. This work acknowledges limitations that future cases must address to examine student learning outcomes. The paper concludes by highlighting the research question: how does GenAI reshape learning as a recursive cybernetic process? Scholars are invited to explore this in their own contexts. 

ABSTRACT. This paper presents a theoretical model for using tactile screens as an active learning pedagogy device to develop critical thinking skills among first-year engineering students, which in turn manifests through the conceptualization of solutions by applying physics concepts, such as the laws of conservation of matter and energy. The model involves the development of an application that allows students to interact with the content of two courses (Application of Conservation Laws in Engineering Systems and Computational Modeling Applying Conservation Laws) using tactile screens with an interconnected interface at Tecnológico de Monterrey, a private higher education institution in northwestern Mexico. By facilitating the comprehension of topics, the change of posture of students (from sitting to standing up), among other advantages, the model 1) intends to develop critical thinking skills through applied examples, and therefore 2) generate a greater diversity of solutions.

ABSTRACT. Threshold concepts are a subset of core concepts that are characterized based on five criterial, troublesome, transformative, integrative, irreversible, and bounded. This paper focuses on the troublesome aspect as a step towards identifying threshold concepts in software engineering and computer science. Survey Responses from 65 participants who reported 109 concepts as being troublesome to them were analyzed and organized hierarchically. The provided explanations were analyzed thematically to identify sources of troublesome knowledge from the students’ perspective.

ABSTRACT. Electromagnetism is a fascinating topic studied in most university physics courses, covering themes such as Lorentz force, Faraday's law, and Maxwell equations. Video-haptic simulators (VHS) are electro-mechanical devices that can simulate forces within a given environment, so that this technology can add the sense of touch to the learning experience. This paper is an extension of a previous article by the authors aimed at measuring the impact of haptic devices in education. We designed, developed, administered, and analyzed a VHS related to understanding magnetic forces acting on conducting wires. The students used Falcon haptic devices in the classroom with a 3D joystick that allows them to feel the force exerted on a simulated particle. Pre-tests and post-tests of basic electrodynamics were administered to experimental and control groups (Ntotal = 236), yielding a relative learning gain difference of ∆g = 0.16 between the experimental group (which used the haptics) and the control group (which did not). A p = 0.045 value for the paired sample indicates that the results are trustworthy within 95%. A perception questionnaire administered to experimental students shows that they consider that the use of the VHS enhanced their understanding of course concepts, thus improving their problem-solving abilities. It is concluded that incorporating external stimulus through touch may increase curiosity in the students and their learning potential.

ABSTRACT. This activity introduces secondary school students to the fundamentals of Agile thinking combined with speculative and futuristic design through a playful, hands-on learning experience. Set in the whimsical world of The Flintstones, the session invites participants to imagine life and education in the prehistoric town of Bedrock, challenging them to collaboratively design a creative, era-appropriate school invention.

The session is structured in four stages: an introduction to Agile principles using real-life analogies, a short science fiction media stimulus (via The Flintstones), a team-based ideation and prototyping challenge, and final two-minute group pitches. The activity aims to foster creativity, collaboration, and systems thinking while offering a light introduction to Agile practices such as visioning, task decomposition, and iterative design.

Empirical data will be collected through pre- and post-session questionnaires, observation of group dynamics, and analysis of student-generated artifacts (sketches and presentations). This study contributes to the exploration of how narrative-driven design and Agile methods can enhance engagement and creative problem-solving among younger learners in educational contexts.

ABSTRACT. Computer simulation and animation (CSA) has received growing attention and wide application in STEM (science, technology, engineering, and mathematics) education. This Work in Progress study aims to assess the effectiveness of an interactive set of CSA learning modules we recently developed to enhance student learning in an undergraduate engineering dynamics course, with a particular focus on particle dynamics. A quasi-experimental quantitative research study was conducted, involving pre- and post-tests administered during treatment and control sessions within an engineering dynamics course. The results show that overall, the CSA learning modules improved student learning gains by up to 30% in the treatment session compared to the control session. Students reported having positive experiences with CSA. It is concluded that CSA can serve as an effective pedagogical tool to supplement classroom lectures and enhance student learning in engineering dynamics.

ABSTRACT. This paper introduces the analyses of the design, create, and pilot four modular courses – Licensing and Open Licensing, Open Education, Open Access and Open Science, and Open Innovation, as integrated in a micro-specialisation of Foundation of Open Practices – to equip students and staff across the E³UDRES² European Universities Alliance with the skills, knowledge, and credentials needed to engage with open practices. The course development methodology involved a collaborative, cross-institutional design process based on the EU DigComp 2.2 Digital Competences framework, Design-Based Research (DBR), intersected with the ABCtoVLE method and refined based on instructional design principles and the Open Educational Practices (OEP) framework. The integration of the Foundations in Open Practices training programme into the E³UDRES² Arena Virtual Learning Environment (VLE), built on Moodle, represents a strategic step toward creating a coherent, scalable, and pedagogically aligned digital learning ecosystem for this courses. The courses contribute to embedding openness within higher education curricula and align with European digital competence frameworks.

ABSTRACT. This paper presents the design, implementation, and evaluation of a GPT-based educational chatbot developed with the Chatbase platform to enhance collaborative work and meaningful learning for students in the Radiation and Radiocommunication course of the Telecommunication Engineering degree at the University of Alcalá (Spain). The chatbot was trained with curated course materials, including lecture notes, laboratory manuals, textbooks, and ITU-R recommendations, and refined through iterative prompt-engineering techniques to improve response accuracy and relevance. Three pilot studies were carried out in real classroom settings involving undergraduate students, comparing the specialized chatbot with general-purpose tools such as ChatGPT and ChatPDF. Quantitative and qualitative data were collected through problem-solving tasks, accuracy analysis, and a perception survey. Results show that the proposed chatbot provided clearer and more targeted explanations, supported students’ reasoning during laboratory tasks, and was perceived as a useful peer-like assistant that saved time and improved motivation. The study highlights the potential of tailored large-language-model-driven chatbots as complementary resources in engineering education, as well as the challenges regarding reasoning with numeric problems and the need for ethical guidelines in classroom adoption

ABSTRACT. This innovation article describes the creation and implementation of LuzIA, a virtual assistant based on artificial intelligence, as a support tool to strengthen self-study among mathematics students in the business area. The project was implemented during the August–December 2024 semester at Tecnológico de Monterrey, within the MA1027 Mathematical Reasoning course, with a control group and an experimental group. The literature indicates that the use of AI, when applied without ethical guidance, can hinder the critical and creative thinking aspects of students’ linguistic and logical-mathematical intelligence. LuzIA challenges this paradigm by being designed as an ally to develop both disciplinary and transversal competencies in students. This report seeks to answer the research question: How can the use of AI foster self-study competencies in university students? The results show that students in the experimental group agreed or strongly agreed with the use of AI, stating that “LuzIA helps improve understanding of mathematics.” An independent samples t-test with unequal variances yielded a statistic of t = 2.071 and p = 0.043, indicating that the difference was statistically significant at a 95% confidence level. LuzIA proved to be an ideal strategy for helping students develop self-study competencies and thereby improve their mathematics performance.

ABSTRACT. This study examines the impact of institutional policies regulating the ethical use of generative artificial intelligence (AI) tools in programming education. Two distinct AI use policies were implemented in a five-week intervention with undergraduate engineering students: one emphasizing reflection through written self-assessment and another emphasizing transparency through the submission of full chat logs. Quantitative data from programming exams and survey responses, combined with qualitative analysis of open-ended feedback, revealed that both approaches led to comparable academic performance, suggesting that the reporting format itself has little effect on learning outcomes. However, differences emerged in students’ attitudes and perceptions. The reflective policy promoted greater ethical awareness, honesty, and self-regulation, while the chatlog policy encouraged broader adoption of AI tools and stricter compliance with reporting requirements. These findings highlight the importance of clear institutional policies that balance flexibility, transparency, and ethical responsibility, positioning AI as a formative aid rather than a substitute for problem-solving. The study underscores the need to integrate ethical and digital literacy into programming curricula and to design educational frameworks that promote responsible and reflective AI use in higher education.

ABSTRACT. This paper is a work-in-progress study that explores how Large Language Models (LLMs) can be used to improve automated short-answer grading (ASAG) in handwritten exams. A hybrid e-assessment approach that combines Handwritten Text Recognition (HTR) and LLM-based grading for Spanish short-answer questions is presented. Initial experiments show that LLMs can accurately transcribe and understand handwritten responses. Commercial models performed best overall, but open-source options also showed promising results. Overall, the findings suggest that this first experience with LLMs could make grading more automated and scalable, helping educators manage assessments more effectively in digital and blended learning environments.

ABSTRACT. We analyzed the ecosystem of the global online STEM program Fab Academy, which graduated 1,697 students in 63 countries between 2009 and 2025 in the Northern (49%) and Southern (51%) Hemispheres. We characterized the profiles of female graduates from Latin America and Africa, regions that gave rise to the Fab Academy-based learning model but show a lower-than-average percentage of women. We identified differences in profiles linked to Architecture and the Built Environment, and Creative Arts (Latin America) and engineering (Africa), with STEM development in the academic context (Latin America) and the industrial context (Africa). Despite its challenges, Fab Lab-based learning demonstrates its impact on women linked to STEM with a model that continues to promote entrepreneurship and management by women in this globalized program.

ABSTRACT. This study presents a successful industry-academia collaboration This study presents a successful industry-academia collaboration between Tecnológico de Monterrey and a leading corrugated packaging company in Mexico, conducted within the Operational Evolution for Industry I and II courses during Fall 2022. The project addressed critical inefficiencies where manual paper combination selection resulted in cost overruns and quality inconsistencies. Historical data analysis revealed actual production costs exceeded targets by 0.85%, with 40.2% of boxes exhibiting Edge Crush Test (ECT) values above specifications and 19.7% below requirements, representing a systematic cost gap of MXN 0.75/m² (6.7% deviation). A multidisciplinary student team developed a comprehensive methodology integrating project-based learning principles with operations research frameworks. The approach included root cause analysis identifying six critical factors and an iteratively refined Mixed-Integer Linear Programming (MILP) model using LINGO software, incorporating paper type, flute profile, target ECT levels, grammage, widths, and real-time inventory availability. Real-world validation demonstrated significant improvements: monthly savings of MXN 1,476,124 (16.13%) for Kraft/white paper families and MXN 1,163,767 (16.45%) for Kraft paper, reducing production costs below targets while improving ECT compliance to within ±5% of target ranges. ROI for the LINGO license (USD 4,995) was achieved within 1-2 weeks, with estimated annual savings exceeding MXN 17.7 million. Beyond industrial impact, this project validated experiential learning methodologies, developing technical competencies (mathematical modeling, optimization software mastery, data analysis) and professional skills (problem diagnosis, stakeholder communication, evidence-based decision-making).

ABSTRACT. This article presents a practical approach to experiential learning (EL) applied to engineering education, utilizing a student wheelchair-building project. The study is theoretically grounded in Kolb's Learning Cycle, analyzing how the project engages its four phases: Concrete Experience (CE), Reflective Observation (RO), Abstract Conceptualization (AC), and Active Experimentation (AE) . The initiative's background is described, highlighting empathy as a key motivator , and results from a survey of 28 students across various engineering programs (Industrial, Mechatronics, etc.) are presented. Findings show how different majors exhibit distinct learning profiles (e.g., Industrial Engineering showed a convergent style ) and validate the positive impact of methodologies like Design Thinking (DT) and Design of Experiments (DOE). Correlation analysis revealed that Abstract Conceptualization (AC) had the strongest relationship (0.78) with perceived learning achieved. The paper concludes that the project is an effective and replicable pedagogical strategy that generates authentic experiential learning by integrating theory, practice, reflection, and action.

ABSTRACT. Empirical investigations into the evolution of student decision-making efficiency within supply chain education remain limited, despite the growing adoption of gamified simulations. This study applies Data Envelopment Analysis (DEA) with a Window Analysis approach to evaluate the supply chain operational efficiency achieved by postgraduate students using Logistics Simulator (LOST) a game-based supply chain simulation during the initial period of an elective course (September cohort). The simulation demonstrated an initial increase in efficiency due to optimized procurement and transportation, followed by significant fluctuations and sharp declines linked to inadequate production planning, ultimately resulting in an average demand compliance rate of only 66.37%. These findings underscore the importance of cultivating anticipatory skills and systems thinking in educational curricula to overcome persistent inefficiencies in cross-functional coordination and production alignment. The study affirms the instrumental role of dynamic simulation environments in fostering critical competencies in resource allocation and risk awareness, essential to modern supply chain management.

ABSTRACT. This paper presents the preliminary results of research on user decisions in a serious game called Logistic Simulator (LOST). We provide an analysis of the first stage of an ethnographic decision tree analysis based on students’ responses. The results show considerable variability in students’ forecasting and production decisions and greater consistency in decisions related to suppliers and raw material orders. The concluding section discusses the impact of these initial findings and outlines the subsequent steps to further develop this research. The results are relevant in two ways. First, the observed decision variability in this study indicates clear pathways to improving the LOST simulator with respect to two types of learning success, and it also reveals the process by which students reach or fall short of them. Our study provides clear tools for teachers to use in the classroom to address student decisions.

ABSTRACT. This working paper integrates a project concerning the use of IA in STEM, and is dedicated to exploring the potential of Artificial Intelligence (AI) in promoting inclusive teaching and learning practices across all scientific fields, with particular attention to STEM. It aims to highlight some of the main findings from the literature in this area and to foster a broader reflection within higher education institutions on the advantages that new AI tools, including chatbots, bring to student engagement – especially for those facing additional challenges (related to gender, social class, ethnicity, or disability). In doing of a diverse and inclusive university for all.

ABSTRACT. Ever since the COVID-19 pandemic, the popularity of e-learning has skyrocketed. A significant challenge facing teachers online is low student engagement, a problem that has been addressed through action recognition techniques. However, these are lagging indicators that only detect after students have already disengaged, allowing little time for teacher intervention.This research develops a novel model that uses action prediction to proactively detect student disengagement. A Sequence-to-Sequence Bidirectional Long Short-Term Memory (Seq2Seq bi-LSTM) model that predicts the future engagement level was built. When presented with only 70 percent of a video it is able to predict the next engagement level with an accuracy of 85 percent. In addition, this paper introduces the first dataset designed for student engagement prediction models. The results demonstrate that action prediction is a better alternative to action recognition for engagement monitoring and should be further researched.

ABSTRACT. Access to real industrial robots is often limited in engineering programs, making it hard for students to connect mathematical models with how real robots actually behave. In this paper, we integrated a remote laboratory --- allowing students to remotely operate an educational industrial robot --- into a course traditionally based on simulation. Using a hybrid approach with pre-test/post-test evaluations, simulator-based modeling, and remote hands-on practice, students first solved a kinematic task in simulation and then repeated it on the physical robot before analyzing differences and adjusting their models. The study showed that this experience greatly improved students’ ability to recognize and reason about real-world behaviors such as vibration, overshoot, and singularities, helping them see how tools like inverse kinematics and Jacobian analysis matter beyond theory. Overall, the proposed remote robot lab proved to be an effective way to bridge the gap between idealized models and practical robotic performance.

ABSTRACT. The flipped classroom methodology combines theoretical and practical learning by encouraging students to work with theoretical content at home and practical in-class activities. The purpose of this study is to provide an active learning approach combining flipped classroom and the incorporation of hands-on experimentation with remote laboratories to improve student engagement and performance, especially in complex subjects or engineering areas, where traditional lecture-based methods are not sufficient. To evaluate the effectiveness of this active learning approach, we conducted a pilot study with a selected group of students, using a combination of predictive and post-experimentation assessments and feedback mechanisms. In the pilot, hands-on experimentation was carried out using our newly developed photovoltaic (PV) solar remote laboratory, to improve the understanding of the influence of altitude on PV energy efficiency, which cannot be achieved with simulations or with a real laboratory deployed in a single location. The results showed a significant improvement in the post-experimentation scores from 50% average to 79% and high student satisfaction with the learning experience. The findings of this pilot study provide valuable insights for the future adoption of similar methodologies and the implementation of the proposed approach.

ABSTRACT. This article presents a Systematic Literature Review aimed at mapping and discussing about methodologies used in the development of computational games for Mathematics Education, highlighting trends, patterns, and gaps in recent literature. The review considered publications from 2019 to 2024, retrieved exclusively from the IEEE Xplore database, using a search strategy structured around three conceptual blocks: game development, mathematics, and methodological processes. After applying inclusion and exclusion criteria, 11 studies composed the final analytical research corpus. The findings reveal a diversity of methods, frameworks, and design approaches, but also recurrent weaknesses, such as incomplete descriptions of development processes, limited integration between pedagogical and computational dimensions, and a predominance of technically driven solutions with insufficient educational grounding. The discussion identifies structural gaps in the documentation of design methodologies and emphasizes the need for more consistent models that articulate pedagogical and technological principles. Overall, this review contributes to mapping the state of the art and offers insights for future research on the development of educational games for Mathematics Education.

ABSTRACT. The rapid evolution of cybersecurity threats needs continuous up-skilling and re-skilling of professionals to maintain strong defenses. However, traditional learning methods often fail to engage students or recommend personalized, adaptive learning paths. This paper proposes a novel knowledge-based recommender system (KBRS) integrated with the 20Q game and reinforcement training. The 20Q game, a decision-making and question-based approach, is adapted to help with upskilling and re-skilling cybersecurity knowledge gaps and recommending tailored learning resources. The system leverages ontologies, knowledge graphs, and expert input to structure cybersecurity knowledge. At the same time, the game engine dynamically adapts questions based on user responses through reinforcement learning techniques that optimize question selection and difficulty progression. Moreover, we incorporate the Learning-to-Explain (LTE) framework, enabling the system to provide translucent explanations for recommendations and question logic, thus enhancing learner understanding. The system aims to enhance engagement, knowledge retention, and skill development by combining gamification with personal recommendations and explanation-aware learning. This research highlights the potential of gamified, knowledge-based recommended systems enhanced with reinforcement training and explanation capabilities to foster continuous learning in cybersecurity, addressing the critical need for up-skilling and re-skilling in this dynamic field.

ABSTRACT. This article analyzes the implementation of Challenge-Based Learning (CBL) in a normatively oriented subject within engineering education. The experience was conducted in the Food Safety Management Systems course at the [Name_University] where students developed the foundational documentation for implementing ISO 22000:2018 and Bolivia’s national SENASAG regulations, in collaboration with the company Bodega del Mar. The study demonstrates how methodological flexibility allows the integration of regulatory rigor with student autonomy and creativity, transforming theoretical learning into an applied process. Results show improvements in the indicators Knowledge, Application, and Professional Values, along with tangible benefits for the company through validated technical deliverables. The discussion highlights the crucial role of the instructor, continuous feedback, and collaboration with real companies as key factors to ensure the effectiveness of CBL in regulatory contexts.

ABSTRACT. This study examines user satisfaction and the integral experience within the educational metaverse "Contiverso," focusing on key dimensions such as perceived utility, technical challenges, and support systems. Using a quantitative cross-sectional design, data were collected from 301 students and teachers through a Likert-scale questionnaire. Results indicate high overall satisfaction (mean=4.21, 73.4% positive responses), with perceived utility (mean=5.31) and avatar customization (mean=4.28) being standout features. However, technical issues like disconnections (56.5%) and slow performance during screen sharing (62.5%) negatively impacted user experience. Support systems, particularly human and tutorial-based assistance, significantly enhanced satisfaction (mean=4.48 vs. 3.79 for low-support users). Regression analysis confirmed the negative effect of technical problems (β=-0.281, p<0.001) and the positive role of support (β=-0.785, p<0.001). The study identifies distinct user profiles based on technical and support experiences, suggesting tailored interventions for optimal engagement. These findings underscore the importance of robust technical infrastructure and accessible support in educational metaverses, offering practical insights for developers and institutions. Limitations include a self-selected sample and the cross-sectional design, calling for future longitudinal and mixed-methods research.

ABSTRACT. In the face of rapidly evolving and increasingly sophisticated malware, traditional cybersecurity defenses are often outpaced. This pa per introduces a novel AI-centric framework that leverages deep learning for the comprehensive analysis, identification, and categorization of ma licious software. By integrating multimodal data streams, our approach harnesses advanced neural architectures to discern complex behavioral patterns and evasive techniques characteristic of modern malware. We demonstrate the efficacy of this artificial intelligence paradigm in en hancing detection accuracy and adaptability against zero-day threats and polymorphic variants. This research underscores the critical role of intelligent systems in fortifying digital perimeters and proactive threat intelligence, offering a robust solution to mitigate pervasive cyber risks in dynamic network environments.

ABSTRACT. Learning thermodynamics involves developing problem-solving skills, and now chatbots can provide valuable support in this process. This study aimed to explore how the chatbot Poe can serve as a guided learning tool for first-year thermodynamics students. This mixed-methods educational innovation, primarily qualitative analysis, measured students' knowledge levels before and after studying the topics of calorimetry, the first law of thermodynamics, and an introduction to waves. Additionally, from a technological perspective, student engagement with the technology was assessed, and a sentiment analysis was conducted. The findings were: 1) student engagement with technology increased, and 2) students had a positive overall experience using the chatbot. These results contribute to pedagogical insights regarding the acquisition of technological skills, specifically the use of chatbots, by engineering students.

ABSTRACT. This paper presents an educational innovation project designed to enhance communicative competence in French as a Foreign Language (FLE) through the integration of Artificial Intelligence (AI) tools. Conducted between November 2024 and May 2025 at PrepaTec Campus Tampico, Tecnológico de Monterrey (Mexico), the study involved 68 students aged 16–18, enrolled in the Multicultural High School Program. These learners, coming from Science, Technology, Engineering, Arts, and Mathematics (STEAM) disciplines, participated in FLE courses at levels 2, 4, and 5, corresponding to CEFR A1–B1. The project integrated applications such as ChatGPT, TecGPT, Hello History, Fliki, HeyGen, Suno, Piclumen, Hailou, and SkillStudio to strengthen oral and written production in digitally enriched learning environments. A mixed-methods approach was applied: quantitative data were obtained through pre- and post-tests (TCF and DELF-style simulations), while qualitative insights emerged from student feedback, audio/video productions, and teacher observations. Results indicate significant gains in fluency, pronunciation, argumentation, accuracy, and learner autonomy, with increased motivation and confidence. This study demonstrates that AI-enhanced environments can foster communicative competence in non-immersive contexts, offering replicable models for integrating emerging technologies into language education.

ABSTRACT. In this paper, the implementation of a living learning methodology in the course “Engineering and Science Modeling” at Tecnológico de Monterrey is presented. This course is distinguished by integrating academic and extracurricular elements, providing flexible, collaborative, and open experiences. As a case study, two activities were carried out according to the methodology, one in a public recreational area and the other in a science fair format. The results of the qualitative/quantitative survey (Likert scale of 1 to 10) demonstrated outstanding success of these activities, with average scores above 8.5/10 and 90% positive comments in the sentiment analysis. Additionally, the correlation analysis validated the central premise of the study by finding a positive and robust association between student satisfaction and their perception of usefulness and the development of professional competencies. In conclusion, this dynamic methodology generated a highly satisfactory learning experience, which was directly correlated with the perceived achievement of academic objectives, confirming its effectiveness in linking students' emotional engagement with learning outcomes.

ABSTRACT. This article presents an international educational innovation experience developed during the February-June 2025 semester within the Global Shared Learning Classroom initiative of Tecnológico de Monterrey. Using the Collaborative Online International Learning (COIL) methodology, students from different semesters and disciplines from Campus Tampico, Mexico, and the University of São Paulo (USP), Brazil, participated. The project focused on analyzing and comparing the performance of a solar panel in two configurations: fixed installation and solar tracking system, aligned with Sustainable Development Goal 7 (SDG 7). Results show that COIL methodology promotes active learning and facilitates the development of sustainable technological solutions with social impact, highlighting the value of the Global Classroom as a catalyst for educational innovation with a human and global commitment.

ABSTRACT. This paper presents the design, implementation, and evaluation of the HydroRocket project, a project-based learning (PBL) experience for first-year engineering students. The activity combined the construction of a water-propelled rocket with digital simulation in MATLAB, data acquisition through Arduino-based sensors, and personalized tutoring supported by artificial intelligence. The primary objective was to enhance engineering design self-efficacy, defined as students’ confidence in their ability to address complex design tasks. The validated Engineering Design Self-Efficacy Scale was administered to 62 students, yielding mean scores ranging from 3.56 to 4.05 with high internal consistency (α = 0.91). Quantitative results were reinforced by qualitative evidence. Additionally, prototype performance improved by 44% over successive iterations, providing objective evidence of design improvement. Findings indicate that the integration of emerging technologies into hands-on projects can foster motivation, autonomous learning, and the development of professional identity in engineering from the earliest stages of education.

ABSTRACT. There has been a well-documented shortage of Cybersecurity professionals in the United States. This deficiency has triggered a well-funded National Cybersecurity education effort both from the part of the National Security Agency (NSA) and Department of Defense (DoD). The most successful Cybersecurity educational movement is highlighted by the NSA through the creation of the National Centers of Academic Excellence (NCAE) in Cybersecurity. There are currently over 400 such designated NCAEs in the United States and all of them have excellent programs to train the next generation of Cybersecurity professionals. Recent changes in the NCAE re-designation criteria revolve around building and assessing Cybersecurity competencies which can be thought of as the ability of a student to perform a task within the context of a work role. In this paper, we explore some efficient ideas and methods in order to allow Cybersecurity NCAEs, who also hold an ABET accreditation in Cybersecurity, to be able to build and assess Cybersecurity competencies in a manner that can be seamlessly used towards ABET program assessment. We hope that NCAE-validated Cybersecurity program coordinators/directors whose program is also ABET-accredited will benefit the most from the ideas and practices shared on this paper.

ABSTRACT. The growing visibility of human identities in online platforms emphasizes the need for creative educational methods to visual privacy and cybersecurity using modern technologies and methodologies. This paper presents a first of its kind generative artificial intelligence (AI) driven visual privacy themed educational game (VPETD), which is an enhanced version of the Visual Privacy Enhancing Technology (VPET) themed game, that was designed to teach the importance of safeguarding visual identities through interactive disguise and detection tasks. Building upon the original VPET game framework, this new version of the game (VPETD) makes innovative use of generative AI, in the form of Stable Diffusion, to create diverse character sets for introducing a novel Doppelganger filter, that enables players to select lookalike disguises using a slider-based interface, and improves the user experience with a redesigned user interface and dynamic image sets. To assess the efficacy of these enhancements, we performed research with a diverse spectrum of participants, including middle school, high school, and university students, as well as educators who played the game as part of different cybersecurity training camps and workshops, or as part of assessment. This paper shares and analyzes all the survey results and the preliminary data collected from these survey responses and demonstrates an increased participant interest in visual privacy and cybersecurity, strong support for curriculum integration, and a high recommendation rate. These findings show that the VPETD game not only enhances awareness of privacy threats but also drives more meaningful engagement with the privacy and security themes by challenging the players furthermore through the unique built-in features. In summary, our findings highlight the potential of gamified, AI-powered teaching tools to bridge the gap between abstract privacy ideas and real-world applications, with implications for wider use in formal and informal learning.

ABSTRACT. Artificial intelligence (AI) is reshaping higher education by enabling instruction to be tailored to individual learners at scale. This paper proposes a conceptual framework for implementing AI-enabled personalized learning in engineering education. The framework integrates adaptive learning algorithms, intelligent tutoring systems, real-time learning analytics, and generative AI tools into a cohesive architecture that personalizes content and support for each student. We review recent literature on AI in education to ground our design in evidence and identify critical considerations for implementation. Key design factors - including scalability, the evolving role of instructors, ethical data use, and equity of access – are analyzed to ensure the framework supports effective and inclusive learning. We clarify the contributions of this work relative to prior studies and avoid unsupported assumptions by citing empirical findings. The discussion addresses practical limitations and challenges (e.g., infrastructure needs, teacher training, algorithmic bias) and outlines future work. This study contributes a structured approach for leveraging AI to enhance engineering and higher education, with the goal of fostering more engaging, adaptive, and equitable learning environments.

ABSTRACT. This paper presents the implementation and evaluation of course-level AI integration across three core courses of a Systems Engineering (SE) specialization within a Master’s program. Conducted as part of a faculty-wide initiative launched in 2024–25, the integration aimed to foster hybrid competencies essential for Industry 5.0. Generative and analytical AI tools were embedded as cognitive partners—aligned with the learning objectives of each course—to support structured inquiry, design exploration, and reflective practice. Drawing on student deliverables, surveys, and comparative assessments, the study demonstrates how scaffolded and context-sensitive AI use enhances both specialized SE competencies and general transferable skills. Findings suggest that aligning AI with disciplinary reasoning promotes deeper learning, increases student self-efficacy, and improves preparedness for AI-augmented professional environments. The paper concludes with instructional design principles to support effective, accountable, and meaningful AI integration in engineering education.

ABSTRACT. This work-in-progress paper evaluates the impact of AI-based teacher-bots (T-Bots) on student performance in higher education. The study was conducted with undergraduate students who interacted with a T-Bot integrated into their learning environment. Using a Technology Acceptance Model (TAM)-based questionnaire and analysis of academic results, we assessed both the perceived usefulness and the actual improvement in student performance. Preliminary findings indicate that students who actively used the T-Bot demonstrated higher engagement and improved academic outcomes compared to previous cohorts. These results suggest that T-Bots can be effective tools for enhancing learning and performance in university settings. Future work will expand the sample size and further investigate the long-term effects of T-Bot adoption.

ABSTRACT. Developing new products is an important process in the food industry, as the success of new launches will largely determine the company's competitiveness and, ultimately, its survival. Creating a successful product involves understanding an increasingly complex context where information is generated at high speed and with elevated density and high dimensionality. That information must be interpreted to define the ever-evolving needs of consumers. In this context, future Food Engineers must develop the skills to utilize the Artificial Intelligence (AI) tools that facilitate acquiring, storing, analyzing, and interpreting the information that guides their food development projects. For this purpose, a case study was developed in which students were tasked with proposing a new product whose commercial viability would be justified by analyzing and interpreting the context of the product category and the consumers' needs. To solve the case, the students would conduct an exhaustive investigation in social networks, scientific articles, databases, etc.; additionally, they would conduct field research using interviews, surveys, and observation. A key component in managing the collected information was the use of ATLAS. ti software, which is commonly used in the workplace, allows, through AI, to collect relevant information from social networks and electronic platforms and analyze this information to find behavior patterns, perform sentiment analysis, and carry out word mining, among others. The study was conducted with a group of seventh-semester students of the Food Engineering program; a group of students from the previous semester, to the implementation served as the control group. The study's premise is that using AI for information management would enable students to understand and interpret the project context more efficiently to make better proposals for new products. The study results showed that students could gather relevant, reliable, sufficient information of different natures and origins, and that, with the help of ATLAS.ti, they stored and organized it efficiently. Students could make deeper analyses of the information and establish more complex relationships between the findings in less time. Students could also represent information and its interrelationships in clearer and more creative ways, allowing them to better understand the context of the development project and generate a proposal for a new food product that is more in line with consumer needs and with greater commercial viability.

ABSTRACT. This study addresses the challenge of efficiently processing Spanish-language qualitative teaching feedback in higher education by developing a systematic approach combining Large Language Models (LLMs) with human oversight. Method: A two-phase procedure was implemented: Phase 1 utilized ChatGPT-4 for generating feedback categories from student comments, followed by expert validation. Phase 2 employed the validated framework to classify individual comments through LLM processing and human refinement. Findings: The procedure yielded eighteen distinct categories encompassing pedagogical and interpersonal dimensions of teaching. The implementation effectively categorizes diverse student feedback while maintaining analytical rigor through human oversight. Originality/Value: This study presents an innovative, resource-efficient approach for Spanish-language institutions to process qualitative feedback, addressing limitations in current NLP solutions while ensuring pedagogical relevance. Research limitations: Current limitations of the procedure include prompt dependency, execution variability, context window restrictions, and the black box nature of LLMs, which impede complete understanding and auditability of the analytical process. Practical implications: The framework enables systematic analysis of qualitative feedback for evidence-based decision-making in faculty development and course improvement initiatives.

ABSTRACT. The objective of this research is to evaluate the level of satisfaction and academic performance associated with the use of the Moodle LMS in 525 students from May to September 2024 of the Basic and Initial Education courses at the Universidad Politécnica Salesiana, Guayaquil. An empirical-analytical approach with a quantitative approach was used, through structured and statistically processed surveys. The results show that accessibility to materials (47 %), evaluation tools (43 %) and additional ICT resources (41 %) were the most valued aspects. 46% of the respondents consider Moodle to be very effective for accessing content, and 50% perceive an improvement in their academic performance. It is concluded that digital platforms are valuable tools, but their positive impact depends on strengthening infrastructure, teacher training and improving educational interaction.