The integration of artificial intelligence (AI) into higher education has sparked increasing interest, particularly in technical disciplines such as building engineering. The present study presents the outcomes of an educational innovation project embedded within a master’s course focused on the intensification and innovation of edification structures. The project aimed to evaluate the efficacy of AI tools in three domains: reference retrieval for academic research, generation of pathological images depicting concrete structure failures, and formulation of sustainable construction solutions using recycled polymer waste. By comparing AI-driven methodologies with traditional manual approaches, the study sought to assess the fidelity, reliability, and technical soundness of AI outputs while exploring their pedagogical value in fostering critical thinking and innovation. The project employed a mixed-methods framework, combining AI survey performance metrics with qualitative evaluations from students and domain experts. For reference retrieval, master’s students utilized AI platforms such as Scite, Elicit, and Consensus alongside conventional database searches (e.g., Scopus, Sciencedirect). The relevance, alignment, and academic rigor of AI-sourced references were compared to those obtained manually, revealing that AI tools demonstrated superior efficiency in curating contextually aligned references, particularly for niche research topics. In the second phase, AI image-generation tools (e.g., Grok, ChatGPT) were tasked with creating visual representations of concrete pathologies (e.g., cracking, spalling, corrosion). These outputs were evaluated against real-world case studies by experts, which identified significant discrepancies. The third pillar focused on AI-generated formulations involving polymer waste in building practices (e.g., plastic-modified concrete). Proposed solutions were scrutinized for technical feasibility, adherence to material science principles, and environmental sustainability, with comparisons drawn to peer-reviewed research and industry standards. AI-proposed material compositions showed promise in ideation but lacked empirical validation. The study highlights AI’s dual role as both an asset and a liability in classrooms. While AI tools enhanced reference retrieval and encouraged “outside-the-box” thinking—particularly in identifying unconventional material applications—their limitations in accuracy and contextual awareness highlighted the enduring necessity of human critical thinking. As such, students often required guidance to discern source or output credibility. In conclusion, AI’s potential to accelerate innovation in structural engineering education is undeniable, particularly in fostering sustainability through recycled material research. However, its current limitations necessitate a cautious, ethically informed approach. The study reaffirms that critical thinking, honed through years of technical training and hands-on experience, remains the cornerstone of pedagogy. By integrating AI as a collaborative—rather than authoritative—tool, educators can cultivate a generation of technical equipped to harness technological advancements without compromising analytical rigor.

Keywords: Artificial Intelligence, building engineering, polymer waste, higher education, sustainable construction.