inside intel? 😉

Excellent investigative journalism Hilbert! Goes to show how a little checking-up can make all the difference...

We expect more info from AMD soon enough as they are the ones to either confirm and/or deny things, we’ll see what they have to say.

THIS is what it's really up to, what AMD has to say to this after they investigated.

Yeh interesting to see how they respond. Regardless, and I've said this before, seems every week some, or maybe thirteen, critical exploits are found. Now. All of a sudden. To the point I don't even read about them anymore. I keep my PC as safe as possible and use it as smartly as possible. Has been enough for 20+ years to avoid exploits. At least that I know of....

This is jut a ruse to put AMD in a bad spot. Its all too convenient specially with the release of Zen+. There are market players that do not want Intel and AMD to get competitive again. Might drive the prices down and God forbid innovation.

It doesn't matter if vulnerabilities are real.

coth:

It doesn't matter if vulnerabilities are real.

Why not? Please elaborate.

So it's ok? If someone finds critical holes in software and hardware - they are safe not to get fixed in case they found by someone who has a tooth?

coth:

So it's ok? If someone finds critical holes in software and hardware - they are safe not to get fixed in case they found by someone who has a tooth?

I'm sorry but I'm not sure I understand your post correctly. In my opinion I'd like to know if those exploits are real, it might very well make a change in my decision to upgrade, or to what product. Hence I'd personally want to know.

Had the exact same feeling when I saw the initial news - starting from name alone. This may sound picky, but 'amdflaws' as a name clearly tries to instigate a negative feeling towards AMD. Professional security firms always stay neutral with regards to the affected companies, like what happened not long ago with Spectre and Meltdown. The lack of technical data in the whitepaper made me doubt further - but what topped it off for me was that these flaws looked like software vulnerabilities, but at the same time they were compared to Meltdown? Doesn't make any sense. And now I see that for the vulnerabilities to actually be exploited you need admin permission on the machine? That's not even remotely comparable to Meltdown. Great article Hilbert, I love the way you pieced the information together.

fantaskarsef:

Why not? Please elaborate.

Not speaking for him, but for one thing the damage has already been done; this will gain momentum regardless of lack of factual basis and it will be an out of control FUDball. We've seen this kind of thing happen many, many times before. Slowing the spread of misinformation is possible, but the fact that it's in the wild means it will be repeated, so it's alive now. And the difficult thing about this type of "crying wolf" is that it's very hard to deny to an absolute. You know that the sun rises in the east, but you can't say it will "forever" rise in the east. There may be vulnerabilities in the architecture that nobody today can ever find, let alone exploit, but the same might not be true five years from now. For all intents and purposes, all these "vulnerabilities" are easily detectable and blockable by security software and are so reliant on too many levels of access being granted to them that it is basically a non-issue; if you leave your car door open on the street, you have worse things to worry about than someone stealing your rearview mirror. (And lets be honest, if you did that, you were kind of asking for it, anyway) The thing about all this I hate the most is there is no real concern here, and it belittles the efforts of security researchers who actually work hard to do their jobs and do it right.

coth:

It doesn't matter if vulnerabilities are real.

You mean vulnerabilities that allows to have your systems compromised to the point wrong doers have elevated administrator privileges with what sounds like literally having physical access to them? If a hacker has gained administrative access to your system or your system is physically exposed to such a person, you have other security concerns to be worried about than the vulnerabilities that may or may not be present within AMD's CPUs. If that is the case, each and every intel system are susceptible to the same issues. In fact, it gets much worse.. people these days can alter intel BIOSes to the point they can manipulate the ME Firmware to support and successfully run hardware which intel themselves have deemed incompatible (Cofee Lake CPUs on Z100/Z200 motherboards, cough) and these tools are found on the internet, freely. What's stopping a master cyber criminal to dupe the average user into flashing UEFI images for their motherboards which promises support for Coffee Lake CPUs while having malware planted in them? Imagine how catastrophic that would be. People are going to download these, flash their systems with them, and will even leave feedback to the OP of the thread, thanking him for his work, while unknowingly leaving their systems compromised. The problem with the PC Community is, we can not act as a unit, and are dysfunctional. We don't look at the bigger picture, all we do is cherish our preferred manufacturers blindly. Like how intel and some nvidia fanboys are rejoicing at this news and making themselves even more stupid by saying things like there's merit to CTS' claims, while in reality all this is an underhanded attack on AMD, and similar things can happen to the brands they cherish in the future. Any piece of electronic is susceptible to vulnerabilities, and nothing is ever full proof. But trust me on this that the findings by CTS are a malicious attempt to sabotage AMD by fear mongering, with even deeper ulterior motives, and nothing else.

First of all: Thank you Hilbert for putting those strange observations together. I already noticed some of them, but not all. But does anyone really give a f*ck about this? Really? I mean you get no "elevation of privileges" out of those. Most (or all?) of them require you to be admin. Did nobody tell anyone it is always bad to have complete admin rights (and probably UAC turned off all together?)? Do you trust and therefore run any program without thinking about it first? To be clear: You have to run the hack as admin. If malware runs as admin without any elevation of privileges before, you are / were already owned / pwned. And think of Intel ME. It is the very same thing as they describe AMD now. You are able to hack Intel ME, therefore there are guides on the net to flash the Intel ME firmware with some dummy one to minimize the risks. And their (amdflaws.com) disclaimer lets the sh!t hit the fan: The report and all statements contained herein are opinions of CTS and are not statements of fact. To the best of our ability and belief, all information contained herein is accurate and reliable, and has been obtained from public sources we believe to be accurate and reliable. Our opinions are held in good faith, and we have based them upon publicly available facts and evidence collected and analyzed, which we set out in our research report to support our opinions. Opinions? Really? "are no statements of facts"???!!! REALLY???!!! Public sources??? Publicly available facts and evidence collected and analyzed??? Do they mean analyzed as in anal? 'Cause this is complete BS they may shovel up their .... you know what I want to say. 🙂 http://cdn.thedailystar.net/sites/default/files/styles/big_2/public/feature/images/plagiarism.jpg EDIT: @Romulus_ut3 : If I could I would like your post a thousand times in a row!!

Independent security researchers have in the meantime confirmed that the exploits are real. And while it does require admin, its not the only exploit that requires a certain degree of access. Being able to easily infest the firmware of the chipset/cpu once you combine it with a privilege escalation attack, you could plant hidden malware which remains unseen for ever. Combining attacks with other attacks is pretty common mode of operation, get entry to the system with exploit A, take it over with exploit B, etc. This is not something you can just ignore.

This is pretty mad. Some insider trading for sure

It has been a day after the news broke on the acclaimed AMD Security Vulnerabilities.

I don't think "acclaimed" is the correct word here, that doesn't make much sense. Maybe just "claimed".

You are totally right, fixed.

nevcairiel:

Independent security researchers have in the meantime confirmed that the exploits are real. And while it does require admin, its not the only exploit that requires a certain degree of access. Being able to easily infest the firmware of the chipset/cpu once you combine it with a privilege escalation attack, you could plant hidden malware which remains unseen for ever. Combining attacks with other attacks is pretty common mode of operation, get entry to the system with exploit A, take it over with exploit B, etc. This is not something you can just ignore.

I think that will certainly need to be addressed. The risks are pretty minimal as you need low-level admin access to begin with, but AMD can't afford to ignore them. We'll see what they have to say.

We'll see about this as more info. is released. At the moment there really isn't enough hard evidence to support either side. I would also like to point out that even if it's done for malicious reasons pointing out legit security flaws with the product is still a good thing long term for the consumer.

nevcairiel:

Independent security researchers have in the meantime confirmed that the exploits are real. And while it does require admin, its not the only exploit that requires a certain degree of access. Being able to easily infest the firmware of the chipset/cpu once you combine it with a privilege escalation attack, you could plant hidden malware which remains unseen for ever. Combining attacks with other attacks is pretty common mode of operation, get entry to the system with exploit A, take it over with exploit B, etc. This is not something you can just ignore.

See, that's the thing: this level of compromised hardware can always be theoretically be possible, given there extreme level of system access privilege. Frankly, I'd be more surprised if a person given complete and total access to a system in this way couldn't insert a way to hook into the system. And any claims of permanence of the firmware-based attacks should always be taken with mountains of salt, because as much as they can tamper the system, they can't lock out original microcode updates. The only real permanent issue with these pieces of hardware is something that fixed-function hardware allows for. (IE, non-flashable execution bits)