Reticulate & Quantropi Talk Synergy & Tackling the ‘Unholy Alliance’ of Quantum AI

Blog

Reticulate & Quantropi Talk Synergy & Tackling the ‘Unholy Alliance’ of Quantum AI

We’re bringing our respective superpowers together to unlock new use cases.

By Anne Wainscott-Sargent

Reticulate Micro just announced an exclusive partnership with Quantropi, an Ottawa cybersecurity firm that has cracked the code on post-quantum encryption by applying quantum science to information security.

Quantropi’s flagship QiSpace™ platform is based on different math that no quantum tech can break (see below). It is being integrated into Reticulate’s VAST™ video compression product and VisionOS™ development platform.

Combining Quantropi’s proven algorithm to Reticulate’s super-fast and super-compressed VAST™ video streaming product results in “a perfect marriage” of aligned partners. On the encoding side or the encryption side, both firms share the same differentiation of working over the tiniest networks with speed, low-battery consumption and minimal to zero overhead.

In the world of tactical satcom, soldiers need to quickly capture and download drone footage even in low-bandwidth environments without it being tracked or manipulated by adversaries. That’s now possible with the two firms integrating their capabilities.

“Because we’re so efficient at math, we don’t need as much CPU time,” explains Michael Redding, chief technology officer for Quantropi. “Giving back CPU time means more battery life for encoding, extending mission time or powering a more powerful sensor. And because it’s zero overhead, it doesn’t slow down. You’re not buffering, waiting for the encryption.”

“Just the encryption algorithm alone, not including VAST, is enough to make a difference on a mission,” says Louis Sutherland, a retired Army major who serves as senior director of business development for tactical markets at Reticulate Micro. “When you pair it with VAST, it becomes the perfect marriage to do both.”

The Three Building Blocks of Cryptography

According to Michael, the power behind Quantropi’s solution is its focus on the three building blocks of cryptography captured by the acronym – TruE – trust, uncertainty and entropy.

“The whole idea is that the beginning of any digital communication, we establish trust. We use specific cryptographic functions to confirm the identity of people we’re talking to in order to establish trust,” explains Michael.

Referred to as asymmetric encryption, this building block often is done through a digital signature. The second part of asymmetric encryption is key exchange, where the parties involved exchange a secret that only they know which is typically utilized as a session key.

Uncertainty is the use of symmetric encryption such as the Advanced Encryption Standard (AES) 256, considered the gold standard as the US government’s recognized encryption method, to protect data being transferred or stored. The third and final building block of cryptography, entropy, involves random number generation to ensure that no one can guess or calculate the number used to create a cryptographic key.

Attacking Cryptography’s Three Legs

These building blocks, much like a three-legged stool, are only as strong as the weakest leg.

“If any of those three things is broken, the chain is broken, the session is compromised. So that gives the bad guys three shots at the title – break the asymmetric encryption, break the symmetric encryption, or break the key,” says Michael, noting that there’s threats against all three with varying degrees of severity.

“The biggest bogey man – quantum – hits most directly on asymmetric encryption: digital signature and key exchange,” he says.
The mathematical formula on how to crack this type of encryption is well documented, adds Louis, “but the compute power to do it doesn’t exist yet.”

For many years, quantum physicists predicted it would be 10 to 20 years before anyone needed to worry. That’s not the case anymore.

A New Science: Quantum AI


“Quantum is moving much faster than anyone guessed. Just last week Microsoft and Quantinuum announced a giant breakthrough that was an 800 times jump forward in quantum computer power,” notes Michael, quoting Gartner Group, E&Y and Cloud Security Alliance who all predicted that quantum will overtake existing encryption methods this decade.

And with the cocktail of quantum combined with machine learning and AI, the window is closing faster. Last month, IBM published a paper co-authored by Asymmetric Encryption luminary Whitfield Diffie that explained that it wasn’t simply machine learning on a classic device or a quantum computer that is the threat, but taking AI and meshing it with quantum computing to create a brand-new science.

And the sense of urgency is here: last week the US thwarted a supply chain attack against a well-known open-source community from an “insider” who had social engineered their way into the community. While it turned out not to be a state actor with a quantum device (which was a specific consideration of the threat analysis), many in the know worry it’s not a matter of if but when that occurs.

Common attacks against the AES-256 encryption standard focus not on breaking it but weakening it, say the two partners. The usual answer in cryptography under threat is to double the length of the cryptographic key against these attacks, but the problem is AES 256 can’t go any higher.

“There isn’t a feasible AES 512 – you can’t just double the key length. It becomes so slow, it literally doesn’t work. It technically can be done but it’s unusable,” explains Michael.

Edge Devices & the Vulnerability of Random Number Generators

The final leg, random numbers, has been a blind spot for a world that continues to depend on personal computers, smart devices and sensors. Even if AES is secure, key generation remains the weak link.

Just how vulnerable is this stool in cryptography, especially with advances in machine learning? According to Michael, if a cybercriminal were to gain access to your Windows PC or Mac 64-bit operating system, using a classic computer running on machine learning, he or she could compromise your random number generator in 37 seconds.

“It’s not rocket science. The problem is millions if not billions of edge devices – your Alexa, your car, your smart doorbell, power grid controllers, even the smart valves the handle your water system that keeps fresh water coming into your house, all are vulnerable. Most of these devices have a brain but not a 64-bit brain, which means the random number generator is garbage. Until two years ago this wasn’t even a threat.”

The problem is also evident on the battlefield where lives are at stake. A report by the Royal United Services Institute for Defence and Security Studies on Russian tactics in the second year of its invasion of Ukraine, found that the Russians were able to break the key generator in Motorola first responder radios used by Ukraine to coordinate troop movements.

And last year, Cisco fell victim to a bad random number generator when Cisco issued 113,000 duplicate digital certificates for their products and devices. The nightmare scenarios is that bad actors could then scan the Pentagon’s public digital certificates for Pentagon network routers, go to Army surplus to find the router’s twin and they would have infiltrated the government’s network.

“That’s a new attack vector – it’s not quantum, but it’s part of the three-legged stool and the unholy alliance of quantum and AI and now thanks to IBM, quantum AI,” says Michael.

How is Reticulate and Quantropi addressing these challenges? Quantropi provides the quantum-secure building blocks through TruE, encompassing trusted entropy, uncertainty, random numbers, symmetric encryption and asymmetric encryption.

“The great news is we’re making it a seamless upgrade,” says Michael, explaining that on the latest demo of VAST and Quantropi, “you just flip the switch and toggle between ‘classic’  AES and QEEP, Quantropi’s symmetric post-quantum encryption, which is up to 18 times faster.”

In fact, a global tactical radio partner benchmarked the speed and reported that Quantropi’s algorithm integrated into software ran five times faster than AES running on hardware. That’s even though hardware nearly always runs faster.

“VAST is super lightweight and fast – delivering streaming performance over the skinniest pipes and on the smallest device, while Quantropi has the fastest encryption with the lightest-weight footprint over the skinniest devices,” explains Michael.

Less power, network and latency, which translates to more energy that can be used for better resolution images and video, or more battery life that translates to longer drone runs.

“In the symmetric encryption world, the big advantage of Quantropi’s solution and the way we use it in VAST is you can turn it on, and the performance will be very close to no encryption at all.” Louis explains. ”At the end of the day, especially in the tactical space, we don’t have time to wait round for things to work in the cloud or the connected environment. This gives us the advantage of working in all those environments.  It makes us so much more secure – it’s the whole security posture we just leapfrogged; we created a new three-legged stool that isn’t vulnerable to current or future ‘unholy’ attacks.”

 

 

Learn More About Quantropi – Enabling Data to be ‘Perfectly Secret’

CTO Michael Redding demystifies the power of Quantropi’s science-based algorithm:

“In cryptography, a perfect form of encryption is called a one-time pad, where you have a secret as big as your data and you only use it one time. It’s called “perfectly secret”. That’s great but if I have a 10-minute video, I need a gigabit of secret. I can’t handshake a gig every time I want to have a Teams meeting.

“We started with quantum physics and said, ‘How do we represent perfect secrecy using quantum state?’ A set of quantum qubits can be used to form a permutation group, which can act to create a quantum permutation pad, which is the quantum equivalent of a one-time pad, so it achieves perfect secrecy. We can represent this quantum permutation group mathematically, and once you have a mathematical expression, you can translate that into computer code that will run on a ‘classic’ CPU.  We basically mapped quantum space into classical space and in doing so, got the power of the quantum effects.”

Copyright © 2024 Reticulate Micro, Inc. | All Rights Reserved.