Transversal Applications of Generative AI in Research and Innovation

Transversal Applications of Generative AI in Research and Innovation Juan José Boté [email protected] Facultat d’Informació i Mitjans Audiovisuals DOI: 10.5281/zenodo.17753856 28/11/2025 Generative AI Generative AI refers to models that create new content: text, code, images. Images Programming source code Generates SPSS scripts Video generation Systems are trained on large volumes of data to detect patterns. They generate outputs on demand, conditioned by our prompts and context. Project Proposals European Projects - INFO DAYS Erasmus+ Capacity Building Horizon 2024 A proposal with GenAi excluded by plagiarism 2025 HORIZON allowed with some conditions Research questions Hypothesis proposal Identify conceptual frameworks Surveys Interviews guidelines Starting point Drafts Planning and Design Generative AI in blank? Formulate a hypothesis Formulate research questions Identify a conceptual framework ?¿ ............... ............... ............... Possible scenarios Planning and Design Generative AI Information ?¿ ............... ............... ............... Data Knowledge Context Possible scenarios Planning and Design Formulate a hypothesis Formulate research questions Identify a conceptual framework DALL·E 3 (OpenAI) – Image generation from text for academic materials. Perplexity AI – AI-powered search engine that synthesizes information from the internet. Canva Magic Write – AI-assisted presentation creation. All these tools can be combined to meet specific research needs. Claude (Anthropic) – Advanced AI model with a focus on security and accuracy. Gemini (Google AI, formerly Bard) – Integration with Google Search and collaborative tools. QuillBot – Rewriting and improvement of academic texts. Grammarly – Spelling and grammar checking with support for academic texts. Copilot (Microsoft/GitHub) – Built-in assistance for writing, data analysis, code generation, and task automation in academic environments. Elicit – Automatic search and extraction of scientific evidence from natural language queries. Connected Papers – Visual exploration of relationships between academic articles. Litmaps – Creation of literature maps for citation and reference management. ResearchRabbit – Search and discovery of related scientific literature. Google Notebook LM – Intelligent assistance for summarizing and exploring your own documents in academic research. GPT-5 with coding functions – Code generation and explanation for data analysis in Python and R. Grok (X / Twitter) – Conversational model for analyzing texts and generating responses in digital environments. DeepSeek – Tools for code generation and multimodal analysis with research support. Orange3* – Data analysis with a visual interface, no coding required. Voyant Tools* – Exploratory text analysis without generative artificial intelligence. Generative artificial intelligence tools useful for academic research. For the writing and revision of academic texts: Per a la recerca bibliogràfica i revisió de literatura: For data and text analysis: For content generation and display: Is the prompt the key to everything? 📌 1. Basic Structure of an Effective Prompt Context: Explain what you are doing (“I am writing an article about…”, “I am preparing a thesis on…”) Clear Instruction: State what you want them to do (“summarize”, “improve this paragraph”, “propose an outline”, etc.) Tone and Register: You can add “in a formal style”, “for a research audience”, “with references if possible” 📋 Example: “Improve this paragraph to make it clearer and more academic, without adding content and maintaining the thirdperson voice.” ⚒ 2. Useful Variations for Research Rewriting: “Rewrite this passage with better syntactic coherence” Organization: “Propose an outline for a scientific article with headings and descriptions” Question Generation: “Give me 3 research questions based on this text” Critical Review: “Identify inconsistencies or gaps in argumentation in this summary” Is the prompt the key to everything? 💡 3. Best Practices Start with a simple request and expand based on the results. Ask for feedback: “What would you improve?” or “Can you offer an alternative version?” If you're not satisfied, rephrase the prompt (not the AI). Objective: To demonstrate live how the same excerpt can be improved with a good prompt. 🧾 Original excerpt (simulated or real, from a thesis or article): “Current research attempts to analyze different aspects of academic communication, but the impact of new technologies on the scientific writing process has not yet been sufficiently studied.” Example 🎯 Prompt 1 (basic): “Improve this text.” 🔄 Response (too generic): perhaps it only makes it shorter or ambiguous. Introduction to data analysis Data collection Boté-Vericad, Juan-José. (2024). ChatGPT como OCR: Generación de ficheros CSV a partir de imágenes con valores de NASDAQ [vídeo] https://hdl.handle.net/2445/214264 -270 Views Boté-Vericad, Juan-José. (2024). Copilot Microsoft como OCR: Generación de un CSV a partir de una imagen con valores del NASDAQ. https://hdl.handle.net/2445/214261 - 300 Views Transcription of interviews or focus groups Cleaning up open-ended responses Optical character recognition Let’s do it Boté-Vericad, Juan-José. (2024). ChatGPT como OCR: Generación de ficheros CSV a partir de imágenes con valores de NASDAQ [vídeo] https://hdl.handle.net/2445/214264 -270 Views Boté-Vericad, Juan-José. (2024). Copilot Microsoft como OCR: Generación de un CSV a partir de una imagen con valores del NASDAQ. https://hdl.handle.net/2445/214261 - 300 Views Excel file analysis Two tabs Data Metadata Let’s do it Boté-Vericad, Juan-José. (2024). ChatGPT como OCR: Generación de ficheros CSV a partir de imágenes con valores de NASDAQ [vídeo] https://hdl.handle.net/2445/214264 -270 Views Boté-Vericad, Juan-José. (2024). Copilot Microsoft como OCR: Generación de un CSV a partir de una imagen con valores del NASDAQ. https://hdl.handle.net/2445/214261 - 300 Views Excel file analysis Two tabs Data Metadata Some tools Dataset Analysis PowerDrill Julius.ai Create reports from datasets Some conditions!!! Make critical and transparent use. Absence of deep reading Cognitive dependence Loss of nuances and context Gerlich, Michael. 2025. “AI Tools in Society: Impacts on Cognitive Offloading and Critical Thinking.” Societies 15 (1): 6. AI Tools in Society: Impacts on Cognitive Offloading and the Future of Critical Thinking - https://doi.org/10.3390/soc15010006 Chen, Youjie, Yingying Wang, Torsten Wüstenberg, et al. 2025. “Effects of Generative Artificial Intelligence on Cognitive Effort and Task Performance: Study Protocol for a Randomized Controlled Experiment among College Students.” Trials 26 (244). https://doi.org/10.1186/s13063-025-08950-3. Resnik, D. B., & Hosseini, M. (2025). Disclosing artificial intelligence use in scientific research and publication: When should disclosure be mandatory, optional, or unnecessary? Accountability in Research, 1–13. https://doi.org/10.1080/08989621.2025.2481949 Checklist for Responsible Use of Generative AI Before using generative AI for a task: When interpreting outputs: When reporting your work Ask whether the task involves confidential, personal or sensitive information. Consider whether non-AI solutions might be more appropriate or safer. Formulate a clear goal and provide sufficient context in my prompt. Treat them as drafts or suggestions, not as verified facts. Cross-check important claims against trusted, primary or scholarly sources. Look for signs of bias, oversimplification or missing perspectives. Be transparent about how AI tools contributed to the process. Assume full responsibility for any errors, omissions or misjudgements. Publishers SAGE https://www-sagepub.com/about/sage-policies/corporate-policies/ai-author-guidelines We believe that AI-assisted writing will become more common as AI tools are increasingly embedded within tools such as Microsoft Word and Google Docs. You are not required to disclose the use of assistive AI tools in your submission, but all content, including AIassisted content, must undergo rigorous human review prior to submission. This is to ensure the content aligns with our standards for quality and authenticity. You are required to inform us of any AI-generated content appearing in your work (including text, images, or translations) when you submit any form of content to Sage or Corwin, including journal articles, manuscripts and book proposals. This will allow the editorial team to make an informed publishing decision regarding your submission. Where we identify published articles or content with undisclosed use of generative AI tools for content generation, we will take appropriate corrective action. Publication ethics policies Artificial intelligence policy Sage recognises the value of artificial intelligence (AI) and its potential to help authors in the research and writing process. Sage welcomes developments in this area to enhance opportunities for generating ideas, accelerating research discovery, synthesising, or analysing findings, polishing language, or structuring a submission. Large language models (LLMs) or Generative AI offer opportunities for acceleration in research and its dissemination. While these opportunities can be transformative, they are unable to replicate human creative and critical thinking. Sage’s policy on the use of AI technology has been developed to assist authors, reviewers and editors to make good judgements about the ethical use of such technology. https://www.sagepub.com/journals/publication-ethics-policies/artificial-intelligence-policy Takeways We learn and use the tools We teach how to use them, question them to the student 3 key ideas Generative AI has become a co-pilot for research in teaching We have examples of AI providing real value AI challenges us as an academic community REFERENCES Gerlich, Michael. 2025. “AI Tools in Society: Impacts on Cognitive Offloading and Critical Thinking.” Societies 15 (1): 6. AI Tools in Society: Impacts on Cognitive Offloading and the Future of Critical Thinking - https://doi.org/10.3390/soc15010006 Chen, Youjie, Yingying Wang, Torsten Wüstenberg, et al. 2025. “Effects of Generative Artificial Intelligence on Cognitive Effort and Task Performance: Study Protocol for a Randomized Controlled Experiment among College Students.” Trials 26 (244). https://doi.org/10.1186/s13063025-08950-3. Resnik, D. B., & Hosseini, M. (2025). Disclosing artificial intelligence use in scientific research and publication: When should disclosure be mandatory, optional, or unnecessary? Accountability in Research, 1–13. https://doi.org/10.1080/08989621.2025.2481949 Boté-Vericad, Juan-José. (2024). ChatGPT como OCR: Generación de ficheros CSV a partir de imágenes con valores de NASDAQ [vídeo] https://hdl.handle.net/2445/214264 Boté-Vericad, Juan-José. (2024). Copilot Microsoft como OCR: Generación de un CSV a partir de una imagen con valores del NASDAQ. https://hdl.handle.net/2445/214261 Boté-Vericad, Juan-José. 2024. “ChatGPT: Descripción de imágenes mediante el chatbot [vídeo].” 30 June. Universitat de Barcelona. https://hdl.handle.net/2445/214260 Boté-Vericad, Juan-josé, Fabeiro, Rosa, Anglada Lara, Ramon. Creando un chatbot con ChatGPT como soporte a la catalogación en bibliotecas, archivos y centros de documentación. Comparación de modelos de lenguaje en versión gratuita y premium. https://hdl.handle.net/2445/213600 Lopezosa, C., Aguilera-Cora, E., Codina, L., & Boté-Vericad, J. J. (2025). Web of Science Research Assistant: Functional analysis and usage recommendations. In J. Guallar, M. Vállez, & A. Ventura-Cisquella (Coords). Digital communication. Trends and good practices (pp. 174-189). Ediciones Profesionales de la Información. https://doi.org/10.3145/cuvicom.13.eng