Ponente: Óscar Díaz (Catedrático de Lenguajes y Sistemas Informáticos de la Universidad del País Vasco).

Lugar: Seminario Mirian Andrés (Edificio CCT).

Hora: viernes 6 de marzo de 2026, 11:00.

Resumen: Engineering education has always aimed to achieve two main goals: introducing students to essential foundational knowledge (like mathematics, natural sciences, and engineering sciences) and preparing them to competently engage in professional practice. Today, however, a significant change makes reaching these goals more challenging—artificial intelligence is transforming engineering practice in key areas such as conceptual design and optimization, simulation and analysis, requirements gathering, documentation and compliance, testing and diagnostics, project management, and risk assessment. 

As a result, teachers are stuck with a tough trade-off: if I lean into AI, am I watering down learning, or if I push back on AI, am I making my course feel outdated? A lot of the worry comes down to this idea that people only really learn wrestling with a problem—and if AI does the heavy lifting, do students actually understand what’s going on? Like, if someone can’t write a SQL query without help, do they really know SQL, or are they just getting used to leaning on a tool?

But there’s a real flip side. Jobs are starting to assume people can work with AI, and if I teach only the old, pre-AI way of doing things, am I setting students up for workflows they won’t use anymore? If the profession has moved forward, it feels wrong for education to stay stuck -students need to know how to use AI well and how to double-check it, not pretend it doesn’t exist.