Hackers are harnessing the techniques of machine learning to extract valuable intellectual property from 3D printed materials according to an article published July 6 in Science Daily. The article reports that a team of researchers from NYU Tandon, including CCS researchers Dr. Nikhil Gupta and Gary Mac, have demonstrated that printer head toolpaths, which hold the unique alignment of composite fibers in 3D printed materials, can be reproduced by hackers using reverse engineering techniques and machine learning algorithms. The researchers found that these toolpaths can be reproduced by training the algorithms over thousands of micro CT scan images. In turn, these algorithms can then predict the fiber orientation on any fiber-reinforced 3D-printed model to a dimensional accuracy within one-third of 1% of the original part.

“Machine learning methods are being used in design of complex parts but, as the study shows, they can be a double-edged sword, making reverse engineering also easier,” Gupta notes in the Science Daily article. “The security concerns should also be a consideration during the design process and unclonable toolpaths should be developed in the future research.”

The study results, which will be published in Composites Science and Technology in September 2020, has also been covered by articles in the July 9 issue of 3D Printing Industry, and the July 13 issue of 3D Print.