Reverse engineering of 3D printed parts by machine learning Reveals security vulnerabilities
NYU Tandon School of EngineeringGlass- and carbon- fiber reinforced composites, whose use in aerospace and other high-performance applications is soaring. Components made of these materials are often 3D printed. Their strength and flexibility depends on how each layer of fibers is deposited by the printer head, whose layer-by-layer orientation is determined by toolpath instricutions in a component's CAD file. A team of NYU Tandon researchers showed that that 3D printing toolpaths are easy to reproduce — and therefore steal — with machine learning. They demonstrated a method of reverse engineering of a 3D-printed glass fiber reinforced polymer filament that, when 3D-printed, has a dimensional accuracy within one-third of 1% of the original part.