Forget about TPM chips for Windows 11, that’s not even half of it

[ad_1]

Microsoft has confused almost everyone with the minimum hardware requirements for Windows 11. At the heart of this confusion is a piece of technology known as the Trusted Platform Module, or TPM.

The task of TPM chips is to perform encryption processes that provide security at the hardware level and verify the authenticity of the system at launch. It also features various mechanisms to make it tamper-resistant.

Among other specifications related to the CPU, RAM, and storage, Windows 11 will require all devices to support TPM 2.0, either built into the CPU or in the form of an additional chip attached to the motherboard.

The announcement sent PC owners to research whether their devices support TPM and, in some cases, how to turn it on in BIOS. The subsequent revelation that Windows 11 can technically be installed on incompatible hardware has added to the confusion.

However, one company is scratching its head over the TPM decision for a different reason. According to Jorge Myszne, founder and CEO of semiconductor startup Kameleon, TPM is already an outdated technology.

“TPM is from 2003; it was good enough 20 years ago, but think about everything that has changed in terms of infrastructure over the past two decades,” he told TechRadar Pro.

“The main challenge is that the TPM is a passive device; while you can store data there and no one can see it, in order to do something with that data the program needs to access it. And if the program has access, the attacker can access too.”

Firmware Security

Founded in 2019 and backed by Xilinx, a leader in programmable SoCs, Kameleon’s goal is to reverse the dynamics of cybercrime, by giving the edge to the defender.

Although the company hasn’t brought a product to market yet, it’s working on a piece of hardware called a Proactive Security Processing Unit (ProSPU) that it hopes can combat the threat of firmware attacks, which are growing in size and complexity.

Myszne explained that “the most common types of attacks take the form of applications targeting higher layers, but they have been blocked with some success.” “As a result, attackers are becoming more specialized, heading down the stack toward firmware; attacks here are harder to detect and persistent.”

Recent figures from Microsoft show that 80% of organizations have experienced at least one firmware attack in the past two years. However, less than a third of security funding is devoted to firmware protection, and 21% of security officials admit that firmware is completely uncensored.

(Image credit: Shutterstock / Gorodenkoff)

The problem with attacks of this type is that they cannot be recognized and blocked by programs. When booting, the system boots in increments, starting with a small area of ​​code loaded into the CPU, followed by a larger set of code, and eventually loading the operating system from the hard drive or network.

“Any compromise that occurs during this process is completely undetectable. The program has not yet been launched, so there is no way to check what is happening,” Myszne noted.

He says the solution to this problem is to have a dedicated device responsible for system security. In the same way that GPUs handle graphics and TPUs handle AI workloads, a security processor is tasked with creating a “root of trust” by verifying that all firmware is genuine.

Dedicated Security Wizard

Designed for servers and data center use cases, Kameleon’s ProSPU aims to address the kinds of problems that arise as a result of TPM chips (and other equivalents) relying on software for instructions.

Whereas TPMs are passive, creating an opportunity for intrusion by an attacker, ProSPU is the master of the system, running active checks to verify that every element of the boot process is genuine. Myszne admits that many of the chips on the market already perform their own secure boot, but there’s nothing there that “walks around in all the different places”.

In addition to establishing this root of trust, ProSPU provides software cryptography services (such as key generation, key management, encryption, and decryption) and runtime security to detect and prevent attacks.

With direct access to memory, under the operating system, ProSPU can operate out of the view of potential hackers. Since it does not rely on APIs for access, there is nothing an attacker can infect it.

“The first thing an attacker does is try to understand the system and the defenses. In this case, the defense is running on a completely different system, with direct access below the software,” Myszne said.

“The attacker does not know what is happening and now needs to attack the system without understanding the defenses. Because the attackers do not like to take risks, they will go elsewhere.”

The future of hardware security

When asked if he thinks Microsoft made the wrong decision by commissioning TPM 2.0 support for Windows 11, Myszne nodded in agreement.

“If I’m working on an enterprise-wide OS, then yes, but for a generic OS like Windows, that’s a big bet, because there will be problems,” he said.

“Usually, TPMs are disabled by default, because they are difficult to manage; you need to know what you are doing or else you risk ruining your computer. How many people know how to tamper with the BIOS safely?”

Although Myszne acknowledges that the TPM is better than nothing from a security perspective, he suggests that the combination of poor user experience and an inadequate level of protection means the requirement will be more difficult than it’s worth.

“The system is not a single chip device as it was 20 years ago. We need the hardware security infrastructure to evolve to meet the needs of today, as well as the needs of the next five to 10 years.”

Kameleon expects an alpha version of ProSPU to arrive by the end of the year, and to be on servers by the first half of 2022. Although applications of the technology are most pressing in data center contexts, given the concentration of risks, Myszne expects ProSPU-style hardware to transition To the consumer, industrial and automotive markets within the next two to three years.

“There is a lot to defend,” he said.

[ad_2]