Tandon professors helped launch “Cryptolets,” a public platform that lets researchers share and implement designs for cybersecurity hardware, making it easier for more scientists to build systems that process encrypted data.
The collaboration between NYU’s Tandon School of Engineering, Stanford University and the City University of New York is funded by a $3.8 million grant from the National Science Foundation. The group originally aimed to develop a library where researchers can share pre-made privacy chip prototypes, combine smaller chips to create more efficient security systems and easily test their designs. Once established, the platform could eliminate the lengthy process of building cybersecurity hardware from scratch.
“Through ‘Cryptolets,’ we will push for a more mainstream understanding and deployment of cryptographic computing,” Tushar Jois, assistant professor of electrical engineering at CUNY, told WSN. “The combination of hardware research infrastructure and new educational materials on cryptography will help those interested in using cryptographic computation but do not immediately have the expertise to build it on their own.”
Cryptography is a type of mathematical modeling that allows senders to scramble information that only the intended receiver can decode. Despite its security benefits, cryptography has yet to see widespread use in computer hardware because it requires expensive chip fabrication and runs significantly slower than other technologies. The “Cryptolets” platform aims to speed up development, allowing researchers to upload chiplets that accelerate specific steps and share code that makes different chiplets work together.
Brandon Reagen, the project’s lead investigator and a Tandon assistant professor, said that he hopes the platform will support cryptography technologies that allow systems to process data without actually seeing it.
“You could, for example, prove who you say you are without actually revealing your driver’s license, social security number or birth certificate,” Reagen said in the September press release.
In a statement to WSN, CUNY computer science professor and participating researcher Rosario Gennaro said the researchers are working to identify the most common steps across cryptographic algorithms so the chiplets they create can be universally applicable.
The platform hosts workshops and competitions — its first was held last week — to gather and publish pre-made encryption chips from students and researchers. The community-building approach will both introduce the platform to more people and vet the systems for quality and everyday use.
“If we can securely compute over medical data, for instance, we can unlock new insights about diseases and treatments without putting individual patient data at risk,” Jois said. “I believe in the importance of privacy, and cryptography is one of our best tools to achieve it in practice.”
Contact Justin Liu at [email protected].