MIT Researchers of CSAIL Release New Fiat Cryptography Tool for Securing Online Communications
Online security is a much bigger issue today than it was in the past, mostly due to several incidents regarding breach of security, hacks and so on and as such, scientists are working tirelessly to create systems that will better secure data over digital spaces.
A major breakthrough has been made in that regard as the Computer Science and Artificial Intelligence Laboratory at the Massachusetts Institute of Technology has successfully created a system that will run mathematical algorithms that will ensure the security of online communications.
This is referred to as fiat cryptography and is it code that secures 90 percent of Google Chrome communications as of now. The paper for fiat cryptography was formally presented at the EEE symposium on security and privacy back in May 2019, though the technology was theorized and deployed as far back as 2018.
How this Works
Fiat cryptography works by generating and optimizing cryptographic algorithms for all hardware platforms which is a process that, in the past, was done entirely by hand but has now become automated.
It is rather interesting that up until a year ago, the Internet data encryption was done entirely by a group of individuals who wrote and rewrote logarithms manually as well as manually weighting the mathematical techniques and chip architectures in order to optimize for performance.
One would have thought such a process would have been automated years ago but obviously, that is not the case.
Obviously, this system had a number of flaws such as human error and as well as non-optimal algorithms. There was also the issue of the maths and chip architecture that the algorithms were made for becoming obsolete and when this took place the process would have to begin manually all over again in order to update it.
Researchers first looked for solutions in C programming and assembly languages after which does techniques were transferred to their code library which is a list of best performing at logarithms.
The use of a compiler for the conversion of programming languages to code well then automatically proofed with Coq, which is a mathematical theorem prover. The best-performing iteration was selected for a particular chip architecture after a number of tests took place.
During the research process, the researchers made use of a body of code that was already in existence and it was discovered that the automated process of generating keys and certificates for data encryption had the same performance as the handwriting techniques but there was less time required to achieve it through automation.
“It’s basically like taking a process that ran in human brains and understanding it well enough to write code that mimics that process,” said Adam Chlipala, a CSAIL researcher who worked on the project.
The fiat photography has since been used by Google's BoringSSL which is an open sore cryptographic library that is used by various programs such as Android apps and Google Chrome.
During his research, Chlipala was aided by several graduate students from CSAIL as well as an engineering graduate student and they are currently working on ways to make the compiler work even faster in searching for optimized algorithms.