Innovation with Design Thinking: Engineering Students vs. Artificial Intelligence in Solution Generation

Resumen

The intersection of human creativity and Artificial Intelligence (AI) in innovation processes is a growing area of exploration. This study investigates the application of Design Thinking (a structured, human-centered methodology) to assess how third-semester engineering students from diverse disciplines (Industrial Engineering, Innovation and Development Engineering, Mechanical Engineering, and Mechatronics Engineering) compare to AI-generated solutions in addressing a real-world educational challenge: designing innovative tools to support children with learning difficulties. A comparative experimental approach was employed, where 67 students, divided into 13 teams, applied the five Design Thinking phases (Empathize, Define, Ideate, Prototype, and Test). Their solutions were systematically analyzed against those generated by AI tools (ChatGPT, Gemini, and Copilot), which followed the same Design Thinking framework. Quantitative metrics, such as the number of ideas generated and prototyping time, were assessed alongside qualitative variables, including originality, feasibility, scalability, and alignment with user needs. Statistical tests (Mann-Whitney U and Student’s t-test) were applied to determine significant differences between human and AI outputs. Results indicate that AI excels in originality, user alignment, and scalability, while students demonstrate greater feasibility and contextual adaptability. AI-generated solutions were consistently limited in number (4-5 ideas), whereas student teams produced a broader range. Additionally, AI significantly reduced prototyping time. These findings suggest that a hybrid approach, integrating AI’s computational power with human-centered problem-solving, could optimize innovation processes in engineering education. Future research should explore AI as a collaborative design tool rather than a competing entity.