Security researchers warned Tuesday that some AMD processors contain "critical" vulnerabilities, as well as backdoors that the researchers claimed were put in place in systems outsourced to a third-party manufacturer by AMD. The 13 different vulnerabilities were found in AMD Secure Processor, which is used in the company’s EPYC and Ryzen CPUs. These bugs would allow hackers who have already gained a foothold into a computer to install persistent and hard-to-detect malware, researchers warned.
CTS Labs, a Tel Aviv-based hardware security company announced the vulnerabilities on a sleek ad hoc website and in videos published Tuesday. The company also published a white paper that explains what the vulnerabilities are without including their full technical details.
Among the most explosive claims in the white paper is the idea that there are “an array of hidden manufacturer backdoors inside AMD's Promontory chipsets” and “the Ryzen and Ryzen Pro chipsets, currently shipping with exploitable backdoors, could not have passed even the most rudimentary white-box security review.”
”This is probably as bad as it gets in the world of security.”
The white paper says that the backdoors were put in place by Taiwanese manufacturer ASMedia, a subsidiary of ASUSTeK, which was recently fined by the FTC for ignoring hardware vulnerabilities. The backdoors “[raise] concerning questions regarding security practices, auditing, and quality controls at AMD,” the researchers wrote in the paper.
“It’s very, very bad. This is probably as bad as it gets in the world of security,” CTS Labs CEO Ido Li On told Motherboard in a phone call.
ASMedia did not immediately respond to a request for comment.
Read more: The Motherboard Guide To Not Getting Hacked
But the claims, and the way they were publicized, have started a controversy in the security community: Some experts say CTS Labs did not give AMD enough time to work on a patch or mitigate the flaws.
All 13 vulnerabilities are exploitable, according to Dan Guido, the founder of security firm Trail of Bits, whose researchers reviewed the flaws and exploit code before publication last week.
“Each of them works as described,” Guido told me in a phone call.
It’s important to note that all these vulnerabilities require hackers to get on the computers and gain administrative privileges some other way first, such as with a phishing attack that tricks the victim into running a malicious application, according to the CTS researchers and Guido.
This means that they are “second stage” vulnerabilities, which would allow attackers to move from computer to computer inside the same network, or install malware directly inside the processor that can’t get detected by security software. This would allow an attacker to spy on the target without detection.
“It makes a bad compromise worse,” Guido said.
Got a tip? You can contact this reporter securely on Signal at +1 917 257 1382, OTR chat at email@example.com, or email firstname.lastname@example.org
Li On explained that one of the vulnerabilities, the one they labelled Ryzenfall 4, can be exploited to install persistent malware directly on the secure processor. Other vulnerabilities could then also be used to escalate into the kernel, the core of the operating system, he added.
That malware “would be very difficult to remove, out of reach of endpoint security solution and basically have full control of the machine,” Li On said.
“The basic nature of some of these vulnerabilities amounts to complete disregard of fundamental security principles,” the researchers wrote in the whitepaper.
CTS Labs said it notified AMD, as well as other companies that use the vulnerable processors for cloud services such as Microsoft, of the vulnerabilities before publicly announcing the flaws. Li On and CTS Labs CTO Yaron Luk-Zilberman said that the researchers sent AMD details of the flaws, including source code for proof-of-concept exploits and more documentation, before publication.
Li On and Luk-Zilberman declined to specify exactly when they notified AMD, only saying it was “very recently.”
“It makes a bad compromise worse.”
Some in the security community criticized the decision of giving AMD such short notice before going public. But the two defended their decision calling it a “public interest disclosure.”
"We are letting the public know of these flaws but we’re not putting out technical details and have no intention of putting out technical details," Luk-Zilberman said on the phone, adding that they have “no intention” of “ever” publishing the full technical details.
AMD did not immediately respond to a request for comment, but told CNET that “security is a top priority and we are continually working to ensure the safety of our users as new risks arise. We are investigating this report, which we just received, to understand the methodology and merit of the findings."
A Microsoft spokesperson said: “We were recently made aware of this report and are reviewing the information.”
These AMD flaws come just three months after security researchers revealed critical bugs in some Intel’s processors, which were called Spectre and Meltdown. Those bugs forced Intel, as well as large cloud providers that rely on Intel-powered servers, to push innovative mitigations and patches that at times hindered processor performance.
Some of these new AMD flaws will be hard to patch, and malicious hackers with a certain level of skill might be able to find ways to exploit them before then, according to Guido. But regular consumers shouldn’t probably worry about them. The real problems, he said, are more for cloud providers and big enterprises.
Get six of our favorite Motherboard stories every day by signing up for our newsletter.