At MIT, artificial intelligence has become so pervasive that researchers often find themselves immersed in it by necessity rather than original design. Sili Deng, an associate professor of mechanical engineering, represents this shift. When the pandemic halted renovations on her combustion kinetics lab in 2019, she pivoted to machine learning to bridge the gap. What began as a workaround evolved into a sophisticated "digital twin"—a virtual replica capable of predicting and controlling fuel combustion systems in real time.
While Deng’s entry into AI was forced by the physical constraints of a global lockdown, others have followed a more collaborative path. Zachary Cordero, an associate professor of aero-astro specializing in aerospace materials, began integrating AI into his work after a cross-departmental introduction. This pattern of interdisciplinary adoption suggests that AI is no longer a standalone field of study but a foundational layer for traditional engineering.
The result is a transformation of how physical systems are designed and managed. From energy-flow devices to novel aerospace structures, the integration of machine learning allows for a level of predictive modeling that was previously unattainable. At MIT, the algorithm has become as essential to the mechanical engineer as the lathe or the wind tunnel.
With reporting from MIT Technology Review.
Source · MIT Technology Review
