AI Agent Architectures: a Comparative Study of Workflows versus A2A (Agent to Agent) and their Application in the Education, Industry and Services Sectorsagénticas de IA: U
DOI:
https://doi.org/10.26439/interfases2025.n022.8430Keywords:
agents, artificial intelligence, education, technology, workflowsAbstract
The development of agentic architectures for artificial intelligence (AI) has led
to new models of interaction and automation, most notably AI agent-based workflows and
agent-to-agent (A2A) communication. This paper presents a comparative study of both
architectures, analyzing their design principles, human-AI coordination capabilities, scalability
and adaptability in different environments, as well as obstacles and challenges. The analysis
shows that while agentic workflows provide efficiency and control in stable environments
with well-defined processes, A2A architectures excel in distributed and heterogeneous
contexts, offering greater flexibility and autonomy. Based on the premise that technology
should strengthen, not replace, human capabilities, this study aims to generate a significant
impact in the areas where it is most needed. Representative examples from the education,
healthcare, and industrial sectors are examined, demonstrating how these architectures
transform key processes. Comparative figures and tables are presented, integrating various
approaches between the two agentic AI architectures. Finally, the technical and ethical
challenges associated with their implementation are discussed, and future lines of research
are proposed for a responsible and effective adoption of these technologies.
Downloads
References
Arslan, S. (2025). Artificial intelligence in food safety and nutrition practices: opportunities and risks. Academia Nutrition and Dietetics, 2(3). https://doi.org/10.20935/AcadNutr7904
Automation Anywhere. (2025). ¿Qué es la automatización robótica de procesos (RPA)? Una guía empresarial. https://www.automationanywhere.com/la/rpa/robotic-process-automation
Balic, N. (2025). Will agents replace us? Perceptions of autonomous multi-agent AI. arXiv. https://doi.org/10.48550/arXiv.2506.02055
Bassey, M., Akpan, J., Ikpe, A., David, V., & Kehinde, T. O. (2025. 26 de agosto). The convergence of additive manufacturing and artificial intelligence with LLMs for smart product design. Academia Materials Science, 2(3). https://doi.org/10.20935/AcadMatSci7868
Binance Academy. (s. f.). Blockchain. https://academy.binance.com/en/glossary/blockchain
Bornet, P., Wirtz, J., & Davenport, T. H., De Cremer, D., Evergreen, B. Fersht, P. Gohel, R. & Khiyara, S. (2025). Agentic artificial intelligence. Harnessing AI agents to reinvent business, work and life. World Scientific Publishing Company.
Casini, L., Marchetti, N., Montanucci, A., Orrù, V., & Roccetti, M. (2023). A human–AI collaboration workflow for archaeological sites detection. Scientific Reports, 13(1), 8699. https://doi.org/10.1038/s41598-023-36015-5
Corvalán, J. G., & Sánchez Caparrós, M. (2025). Agentes de inteligencia artificial y workflows agénticos: la nueva frontera de la automatización. Laboratorio de Inteligencia Artificial de la Facultad de Derecho UBA. https://www.web.ialab.com.ar/webia/wp-content/uploads/2025/libros/Agentes-de-inteligencia-artificial-y-workflows-agenticos/Agentes-de-inteligencia-artificial-y-workflows-agenticos.pdf
Coshow, T. (2024, 1 de octubre). Los agentes inteligentes de IA pueden operar de forma autónoma. Gartner. https://www.gartner.es/es/articulos/agente-inteligente-de-ia
Empresa Actual. (2025, 11 de febrero). ¿Qué son los agentes de IA o la inteligencia artificial agéntica (IAA)? EmpresaActual.com https://www.empresaactual.com/que-son-los-agentes-de-ia-o-la-inteligencia-artificial-agentica-iaa/
Hauser, E. (2019, 9-13 de noviembre). AI systems as ethical agents [Presentación dentro del taller Good Systems: Ethical AI]. 22nd ACM Conference on Computer-Supported Cooperative Work and Social Computing (CSCW), Austin, Texas, Estados Unidos.
Li, Y. H., Chang, Z. H., Shan Ou, Y., Chen, Y., Yu, L. Y., Tan, J. Q., Liu, Y., & Chang, Y. Z. (2025). Exploration of the development and technical features of intelligent inspection robots. International Journal of Advanced Engineering Research and Science, 12(10), 61-70. https://doi.org/10.22161/ijaers.1210.9
Lieben, M. (2025). 3 types of AI workflows [flujograma]. LinkedIn. https://www.linkedin.com/posts/the-best-of-ai_ai-agents-are-replacing-human-employees-activity-7364786657318838273-SH--/
Ramalingam, B. C. (2025). Efficient implementation of AI agents in enterprise application integration (EAI) and electronic data interchange (EDI). International Journal of Scientific Research in Computer Science, Engineering and Information Technology, 11(2), 150-170. https://doi.org/10.32628/cseit251112397
Samdani, G., Viswanathan, G., & Dasu Jegadeesh, A. (2025). Human-AI collaboration: Balancing agentic AI and autonomy in hybrid systems. International Journal on Cloud Computing: Services and Architecture, 15(1), 1-15. https://doi.org/10.5121/ijccsa.2025.15101
Shaw, F. X. (2025, 19 de mayo). Microsoft Build 2025: The age of AI agents and building the open agentic web. Official Microsoft Blog. https://blogs.microsoft.com/blog/2025/05/19/microsoft-build-2025-the-age-of-ai-agents-and-building-the-open-agentic-web/
Shoham, Y., & Leyton-Brown, K. (2008). Multiagent systems. Algorithmic, game-theoretic, and logical foundations. Cambridge University Press.
The Best of AI. (2025). AI agents are replacing human employees. Build yours [Publicación]. LinkedIn. https://www.linkedin.com/posts/the-best-of-ai_ai-agents-are-replacing-human-employees-activity-7364786657318838273-SH--/
Viswanathan, G., Samdani, G., & Dixit, Y. (2025). AI Agents. International Journal of Advanced Information Technology, 15(1/2), 9-17. https://doi.org/10.5121/ijait.2025.15202
Xu, Z., & King, I. (2025, 2 de septiembre). Introducing Academia AI and Applications: A new platform for responsible and interdisciplinary AI research. Academia AI and Applications. https://doi.org/10.20935/AcadAI7885
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under an Attribution 4.0 International (CC BY 4.0) License. that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
Last updated 03/05/21
