An intriguing paper published by researchers at the University of Michigan describes a new processor architecture capable of self-encryption that can fend off any hacks.
A DARPA-supported project, the new chip design renders the current electronic security model of bugs obsolete – or so claim its creators.
“Called MORPHEUS, the chip blocks potential attacks by encrypting and randomly reshuffling key bits of its own code and data 20 times per second—infinitely faster than a human hacker can work and thousands of times faster than even the fastest electronic hacking techniques,” according to the press release.
Professor Todd Austin, one of the co-creators, likens the chip to a “Rubik’s Cube that rearranges itself every time you blink.”
The processor is still in prototype phase, but its creators claim it has successfully defended against every known variant of control-flow attack, a widely used technique by attackers.
The technology can be employed by virtually every type of device that boasts a CPU, from laptops and PCs to Internet of Things devices and even smartphones.
“We’ve all seen how damaging an attack can be when it hits a computer that’s sitting on your desk,” Austin said. “But attacks on the computer in your car, in your smart lock or even in your body could place users at even greater risk.”
A hacker would find it impossible to pin down any vulnerabilities, the paper says, because it constantly randomizes critical program assets in a process known as “churn.”
Specifically, it randomizes bits of data known as “undefined semantics” with a churn rate of once every 50 milliseconds. The performance compromise is just 1%, the paper authors claim.
Prof. Austin is already pushing for the commercialization of MORPHEUS via a startup called Agita Labs, co-founded by himself and colleague Valeria Bertacco. Geeks can access the full paper here: MORPHEUS: A Vulnerability-Tolerant Secure Architecture Based on Ensembles of Moving Target Defenses with Churn.