Microsoft BitLocker – YellowKey zero-day exploit
tomshardware.com
tomshardware.com
What will it take for more companies to truly understand their risks with Windows and being locked into Microsoft’s platforms?
If this is a fed mandated backdoor, I guarantee Microsoft/Windows isn't the only one either, they are just the only/first ones to get caught. I'd be suspicious about every single commercial, closed source operating system or encryption product in the US right now.
That's what this is about. Microsoft doing bad security practices while trying to get away with it, leading to this outcome.
The researcher also claims to have another version ready which allows to also bypass TPM+PIN via a similar backdoor, which I'm inclined to believe.
Why do I believe that? 5 ring 0 zero days within 3 months are so statistically unlikely to be found, by the same person, in such a short time. Whoever this person is really knows their exploits, and must be in the league of Juan Sacco.
so I call bullshit on the PIN bypass
A USB stick containing a masterkey to decrypt a bitlocker volume is literally the definition of a backdoor.
Go on, try it out. It works.
thats an LPE, not an encryption backdoor
the USB stick doesnt decrypt bitlocker, it just gives you root after bitlocker was AUTOMATICALLY decrypted
> thats an LPE, not an encryption backdoor
No. RedSun and Bluehammer were LPEs
> the USB stick doesnt decrypt bitlocker, it just gives you root after bitlocker was AUTOMATICALLY decrypted
No, that's not what the bypass does. Maybe go try it out and verify it before you come to your quickly made conclusions?
It's not tied to "automatically decrypted" volumes, whatever that would imply for your setup requiring a pretty pointless TPM keystore for that.
If your case were true, it would also imply that any bitlocker cryptography never really worked because it was automatically decryptable without the need for a password/hash/whatever to get your keys from the keystore, which actually makes it so much worse. Even worse than the previously known coldboot attacks.
How could anybody besides a Microsoft employee, given the appearance of this bypass technique?
In the default BitLocker configuration, Windows puts all the key material in the TPM, locked behind the usual trusted-boot stuff: known-good BIOS hashes the bootloader and tells the TPM, bootloader hashes the kernel and tells the TPM, kernel hashes the initial process and tells the TPM, (I’m not sure how far it goes in this specific application,) and at the end of it the TPM won’t release the keys unless the entire chain was correct. This process does (modulo TPM flaws) ensure the disk will only be decryptable when in the original computer running the original OS. It does not ensure that the original OS will not subsequently give a root shell to anyone who walks up to the keyboard and types in a cheat code, and that’s essentially what’s happening here.
Celebrite et al. take a similar approach: after your Android phone boots and you first enter your PIN (which, unlike with BitLocker defaults, is required to unlock the TPM, thus the distinguished status of “before first unlock” aka BFU vs “after first unlock” aka AFU), the key material is already in RAM and breaking dm-crypt is not necessary; all that’s needed is find a USB stack vulnerability or a Bluetooth stack vulnerability or whatnot that can be leveraged into a root shell.
Agreed that the default Bitlocker config is much less secure than having a PIN at boot time due to the amount of code that gets run.
Someone else claimed this doesn't affect people who actually care about security and enable boot-time password protection.
https://post-cyberlabs.github.io/Offensive-security-publicat...
https://blog.scrt.ch/2024/10/28/privilege-escalation-through...
Yes, the PIN is entangled with the key material.
So true.
This means it's vulnerable to an offline bruteforce attack to derive the PIN.
So it's still doable, even in an automated fashion, just slower.
With today's multi-GPU cloud systems available to everyone with a credit card, you can probably crack the default-length 6-digit PIN the same day you extract the key protector.
> The article shows that the PIN-entangled key material can still be downloaded directly from the TPM.
Not exactly, the TPM has PolicyAuthValue(PIN), so the PIN also needs to be provided to the TPM to unseal the material, and the hardware anti-hammering should prevent brute forcing it this way. The blog post documents dumping the PIN-entangled key material by MITM-ing the TPM communication while a user enters the PIN; the entanglement is a belt-and-suspenders approach.
Where are you seeing that? I can't find it in the article.
It wouldn't make sense to me for that to be the case if the article details how the driver does it own unwrapping/decryption after the KP is extracted. Plus it would probably mean they're lying about TPM+PIN being defeatable.
> The blog post documents dumping the PIN-entangled key material by MITM-ing the TPM communication while a user enters the PIN
I really don't think so... the screenshot with the PIN entry I think was only for hooking his debugger up in order to reverse the driver's decryption process. I don't see where they mention how/when the KP is actually extracted. It looks to me like it's transmitted during boot _before_ the PIN entry, so that the software driver can decrypt it after the user enters the PIN.
They list the steps as:
1. Extract TPM data. The TPM data is encrypted Key Protector (aka KP).
2. Generate the decryption key of KP
3. Decrypt KP
4. Extracting encrypted VMK
5. Decrypt VMK using KP
I didn't see anything about needing to enter a PIN in order to get the TPM data.
If the TPM required a PIN to extract anything, I think there would be no need to manually decrypt anything in software as they show with the python code.
Of course I could be wrong... please feel free to provide more info.
Like I specifically pointed out, it's belt and suspenders.
> Of course I could be wrong... please feel free to provide more info.
From https://blog.scrt.ch/2024/10/28/privilege-escalation-through... :
> Indeed, by analysing the decryption process, it appears that the user’s PIN is sent to the TPM which releases the intermediate key only if the provided secret is correct, thus effectively preventing offline bruteforce attacks.
> Secondly, no data is returned when the PIN is incorrect, which indicates that the PIN or a derivative is sent to the TPM for verification.
Now I have to wonder if the exploit author's definition of "it works with a PIN" is simply "it works if you enter the correct PIN" and just somehow left out that important detail... I don't know. Perhaps everyone is just guessing that they meant it's possible to exploit without knowing the PIN at all.
I suppose they could be lying too, but I would hope they would be smarter than that given their apparently successful track record /shrug
Belt and suspenders = the industry standard term for "you have one protection you rely on, you add a second that should help." Stuff like ASLR, for example. Or in this case, the stretched key material. The belt is the TPM PIN anti-hammering, the suspenders are the key stretch / entanglement.
> Perhaps everyone is just guessing that they meant it's possible to exploit without knowing the PIN at all. I suppose they could be lying too, but I would hope they would be smarter than that given their apparently successful track record /shrug
Trusting the word of exploit developers, especially random anime avatars on GitHub, is always a bad idea no matter the recent track record. Self promotion is very powerful in the security industry and every claim deserves independent research; that's at least half of the original point I was trying to make about conspiracy theories.
Personally, I suspect the exploit author had a disk with multiple enrollments in addition to the TPM + PIN one, and broke a parallel strategy.
I don't remember which updates triggered it, but that was September 2015.
From the perspective of the TPM, I have now learned that it is required for it to release the key.
Perhaps those updates didn't really reboot in the traditional sense. If you turn the machine fully off and then back on, and it still doesn't ask for a PIN... now you have my attention.
Back in the windows 7 days you could stick a windows installer CD in and press Shift+F7 or something and get a system command prompt with the drive unlocked.
Surely when someone said 'we're gonna let the installer unlock bitlocker' they immediately thought 'That means the whole installer needs to be as secure as the login screen' right? Seemingly not.
On my birthday iirc once long time ago I think in 5-6th class not sure, my brother gave me his laptop, I wanted to do python but python wanted admin password on windows to install properly. So what I did was I dont even remember how, but download one operating system which could then crack the windows password so that I can set new and I used that to then set a new password to then install python. to then only print hello world :D (I think only because one of the cousins I really admire mentioned that he made 2k loc of python once and I thought during that time, python is the endgame). We are talking about windows 7 but I think that windows 10 security must've gotten better. So these are some things that I have done, I wouldn't call it coding as much as tinkering but I love doing these things from as long as I can remember :D
I just remembered this paragraph from one of the comments that I had written sometime ago, source: https://news.ycombinator.com/item?id=47663383
Which really annoyed me. Desktops don't need encrypted drives.
- The feds show up
- A bugular breaks in and grabs your computer
- You're selling your house and host an open house
- You have curious children and want to keep them from live booting and reading your tax returns
What? Most Linux distributions don't even enable FDE by default, and even when they do, they frequently use the exact same system as BitLocker (automated unlock sealed to TPM PCRs) with the exact same vulnerabilities (any signed OS image with a postboot authentication bypass gets you the disk content, as does any method for inspecting the state of system memory). This is an architectural tradeoff you can make on any platform and has nothing to do with "lock in."
It's straightforward to configure BitLocker disk encryption to be more secure, but it creates enormous headaches for admins, so they don't do it.
I do think that Apple have some better security defaults for FileVault ("enabling" FileVault basically consists of wrapping the existing hardware UID entangled key with the user's password as well) but this strategy does create big issues with remote password rotation or delegated authentication (ie, Active Directory), which is probably why Microsoft don't choose it as a default.
Do they? Any time I've done FDE it's always been luks with a password, I've never seen one go for TPM by default!
I've only recently implemented luks+TPM on a personal laptop (and that was a PITA to do).
I didn’t find it too difficult to set up TPM backed encryption on Arch using systemd-cryptenroll for my home server, although for anything I use interactively I just use a passphrase instead.
It also strikes me that these are several very high value (all but one complete) exploits.
Surely the value of these on the market would be astronomical and best suited to law enforcement agencies using unlock as a service businesses.
So I have to say I applaud the open disclosure
They also seem suitable for using a secret sharing scheme.
I have Microsoft authenticator requests all day every day. Using aliases has helped but somehow they continue. It's only a matter of time before somehow accidentally I approve.
Which has simply led to me not putting anything of high value in my Microsoft account and not using it for my email.
Local accounts still work fine for Win 11 Pro, I installed it a few days ago using a clean ISO directly from Microsoft. No special patching or command line stuff needed, making a local account is part of the official install process.
https://www.xda-developers.com/microsoft-cracking-down-on-lo...
https://medium.com/@michaelswengel/microsoft-blocking-local-...
I had someone demo me preserving PCR values by patching SMM module in firmware without triggering any bitlocker lockout, this also means that you can externally write bios with the smm module as long as you have ~2 minutes to disassemble the laptop or desktop and flash firmware.
This hurts the most when you don't have PIN authentication which means you just need to steal the laptop to exfiltrate data, if you do then you have to have the user boot which then drops a payload exfiltrating data over network or just stealing the laptop again as you can write back decryption keys into non encrypted partition or corrupt some sectors at the end of the disk and write them there.
* modifying smm allows you to patch the boot process loading a malicious payload into hypervisor/kernel.
Majority of hard disk encryption done in the HDD/SSD controller is 100 times more crap than BitLocker itself. It's littered with bugs and security vulns. Anybody using it is insane.
you use veracrypt which doesn't have any hardware attestation (convenience) features, but it does still leave you vulnerable to the same surface PIN+TPM is vulnerable to. the real defense is making it so opening your laptop/desktop physically fuses something via latch and wipes the key off your system requiring re-entry.
of course, who wants to own a laptop/desktop that you can't open we have enough of that with our phones.
Less emotionally but mostly equivalently, the expense and non-cryptographic skill requirements of breaking mass-storage crypto are quite high while the rewards are comparable to those from breaking much softer targets, so the absence of results since that one paper only changes my mind very slightly. Besides, we know plenty of examples of what these kinds of opaque, serious-business, pay-to-play environments produce: cellular crypto is an uninterrupted series of disasters, so is Wi-Fi, and the things that we do know about storage devices don’t point to an outstanding culture of cryptographic competence there either. Once you’ve done enough to slap an “OPAL” label on it (which says nothing about the internals), there’s just no competitive pressure to improve.
There is a right way to do all this, and it’s essentially what NICs do: allow the host to offload symmetric crypto to the device, but keep the results of said crypto accessible at any moment. And it’s not like there are even that many modes used in full-disk encryption, let alone ciphers.
It’s also a way of saying that the entire approach taken by hardware disk encryption (unspecified crypto done inside the device in an unverifiable manner) has, with the benefit of hindsight, proven fundamentally flawed despite its reasonable appearance (in every system which had used it, not just storage), and I wish there was a way to pressure (consumer) storage vendors into going in a different direction. It is simply never a wise choice to trust people’s opaque crypto, however competent they are.
Oversimplified and not accurate. Some manufacturers had flawed implementations, others did not. Also, that was a long time ago. There are advantages to hardware encryption. It preserves performance and mitigates other vectors like cold-boot attacks without having to encrypt RAM, which also comes with a performance penalty. By the way, both software and hardware-based encryption can be combined. Cryptsetup on Linux actually offers this, and before you ask, the keys are split. If one is compromised, the other remains secure.
I know my bike lock can be cut within seconds by someone who is sufficiently skilled and determined. I'm still going to lock my bike.
Also, I'd argue the stolen bag scenario usually features dilletante attackers.
> (Note: The YellowKey author disagrees that PIN is a protection
God I hate this stupid design of burying the decryption key in the TPM and hoping the software does not get fooled to reveal it.
Microsoft always sucks. Why don’t you ask for the password at boot time and derive the key from it. So much simpler and makes this kind of attacks impossible. Nobody is going to bypass LUKS or FileVault like this.
Does anyone really trust these shitty Windows laptop/desktop manufacturers to get these things right? These guys couldn't even get basic hardware features like trackpad drivers right.
author could become famous by being the first to proove an actual backdoor in an OS disk encryption
That's enough proof.
Since there's a ton of misunderstanding in this thread, I'm going to go into how disk encryption works conceptually.
First, there's a symmetric key to encrypt blocks on the disk. Since you want to be able to change your unlocking password/mechanism without re-encrypting everything on the disk, this has nothing to do with unlocking the disk. This is what you want to get BY unlocking the disk. Let's call this the "data encryption key".
Then, there's something you use to encrypt the data encryption key. Let's call this the "key encryption key" (abbreviated KEK from here on in).
When you use a TPM, the KEK is stored inside the TPM. When you use a TPM PIN, the TPM refuses to release the KEK for use by the OS unless that PIN is provided.
You could say "why not make the KEK be a hash-mixed combination of a PIN and something inside the TPM?". One could do that! But that's not how Bitlocker works. There is a reason it doesn't work that way: the TPM is supposed to let company admins in charge of the device access it even if the original PIN is forgotten, by using other policies letting them get at the KEK. I personally set my own devices up such that the passphrase IS part of the KEK itself.
Interestingly, LUKS does not have a composite key mode natively that lets you combine a password with TPM material, but there are some good reasons not to use JUST a password:
1. The strength of your disk encryption reduces to the strength of the password, where a TPM can have a 256-bit truly random key
2. If someone keylogs the password, or tricks you into disclosing it, they can later decrypt your drive from anywhere, where a TPM binds the attack to those with posession of the TPM
3. There is no protection against brute force attacks (rate limiting), where a TPM does - or tries to - impose a rate limit
Now, let's go on to what YellowKey attacks.
A TPM can have inside itself "registers", called PCRs. These PCRs can be updated but not reset - think of it like you can add numbers to them but not subtract, and they only go back to zero when you reboot.
Using a passwordless encrypted boot, the TPM is configured to only release the key when the PCRs are in the exact correct state. As the OS boots it adds numbers to those PCRs. If you boot "the wrong" software, the numbers in those registers won't match the expectations, and you cannot unlock the disk.
Speculation on my part: the reason there's an exploit here is that the Windows Recovery Environment apparently can match the PCR values for the booted OS, causing the TPM to release the key, but WinRE doesn't require you to get your password right before it gives you access to the data. So far as I know, protecting the TPM key with a PIN would mitigate this issue, but it's still bad.
Or maybe the exploit actually does something inside the TPM itself, causing it to unconditionally release the key even when protected by a PIN: that would be even worse, but **NOT*** a problem with Windows. That would be a problem with the TPM.
For example, storing a second, hidden copy of the master data encryption key, in an obfuscated form on a region of the disk that is unused or somehow reserved for the OS.
An un-exploited system is booted with a modified version of the Windows Recovery Environment.
Like I said, I think the not-well-described problem here is that (effectively) the lock screen on Windows RE is not secure, so you can have a PCR match in the TPM, but then access the disk as an administrator without typing the admin's user account password. That's not a vulnerability of the TPM itself, and it's not some kind of persistent exploit. It's a flaw in the Windows RE.
I'll also point out it grants access to do only what Microsoft themselves could do at any point. Anyone who has the ability to make a validly-signed copy of Windows could break into a TPM-locked Bitlocker setup exactly this way. People who use Bitlocker without a PIN are implicitly accepting that risk.
So this only affects a particular mode of bitlocker in which the drive is automatically decrypted on boot without the user providing any secret. Meaning the key is basically stored in plaintext on-device, albeit in a convoluted way.
To me it seems intuitive that such a mode isn't secure. It's a bit like protecting your door with an unpickable unbreakable lock, but then putting the key in a lockbox on the wall with a flimsy padlock that can be raked or cut off in seconds.
It seems roughly equivalent to not encrypting the drive at all so it doesn't seem surprising that there's a way to bypass it.
Encrypting the disk using a secret password provides advantages over encrypting the disk using a public password.
Encrypting the disk using a public password again provides advantages over not encrypting the disk (such as being able to securely "delete" data by removing the data encryption key).
I agree with your core point that attempting to use measured boot and secure boot to control whether the disk can be decrypted is full of holes. But if you want the computer to have an encrypted drive and to be able to boot up without a network or human intervention, what are your options really?
If the TPM is properly designed and manufactured, and the software relying on it is again properly designed and implemented, then it would be perfectly secure. The problem is more the difference between the theory and the real world; the flimsy lockbox analogy doesn't hold.
I think they're attacks on Windows' measured boot approach.
True. It's unfortunate, amd a lot of false information being spread there.
> the KEK is stored inside the TPM
That's not how it works. The KEK is not stored inside the TPM, but encrypted/decrypted by the TPM.
> You could say "why not make the KEK be a hash-mixed combination of a PIN and something inside the TPM?".
Bitlocker does that. Cryptenroll doesn't (https://github.com/systemd/systemd/pull/27502), which is bad but has not been fixed.
TPMs are a nice idea, but there are a few problems:
- The KEK should also depend on the PIN. Cryptpentroll does not do this at all and Bitlocker limits the PIN to 20 characters.
- There are various manufacturers of TPMs and all of them have different implementations. Some of them had been broken in the past, which is why it's important to make secrets PIN-dependent.
I seriously doubt the author found a way to bypass PIN protected setups in general. This should only be possible in combination with a vendor/model specific vulnerability. Maybe an fTPM?
As of this moment, I would rather look at it as a convenience feature. A high entropy password + a proper KDF (not possible on Windows) like scrypt or argon2 is the better choice. Encryption should be handled by SoC engines like on Macs, iPhones or some Android phones to mitigate other attacks and preserve performance anyway. Panther Lake CPUs with vPro support do on Windows.
No, the KEK is stored inside the TPM, and the DEK is decrypted by it. If you have three layers (two KEKs, one encrypted with the other) then sure. But at the end of the day one private key is actually stored inside the TPM, taking up as much space as the length of the key.
To be specific, Bitlocker uses a stored key, not a derived key.
> Bitlocker does that. Cryptenroll doesn't (https://github.com/systemd/systemd/pull/27502), which is bad but has not been fixed.
I don't think Bitlocker does that, because it's possible to set up admin PIN recovery techniques. That would be impossible if the PIN were part of the KEK.
> - The KEK should also depend on the PIN. Cryptpentroll does not do this at all and Bitlocker limits the PIN to 20 characters.
I agree.
> There are various manufacturers of TPMs and all of them have different implementations. Some of them had been broken in the past, which is why it's important to make secrets PIN-dependent.
I agree. Perhaps consider using a FIDO2 token (supported by cryptsetup) instead of a TPM. There are open-source implementations of FIDO2 and open-hardware ones too. There are even open-source implementations of FIDO2 where the key is in fact derived from the user's PIN (plus a stored secret). If you did that, you get the proper security properties even without cryptsetup mixing the two methods.
> I seriously doubt the author found a way to bypass PIN protected setups in general.
I agree - I think the author probably just found a flaw in the windows recovery environment and is talking up how a PIN only helps you if the attacker does not know your PIN (in other words, acting as if the PIN provides no threat resistance when really it provides an additional layer).
> As of this moment, I would rather look at it as a convenience feature. A high entropy password + a proper KDF (not possible on Windows) like scrypt or argon2 is the better choice. Encryption should be handled by SoC engines like on Macs, iPhones or some Android phones to mitigate other attacks and preserve performance anyway. Panther Lake CPUs with vPro support do on Windows.
I think the best you can do right now is to layer a password with a hardware device. I don't think saying that the hardware devices are flawed means they are not useful as PART OF the security setup. It would certainly be nice if the software did this automatically/easily and it's unclear to me why it does not.
https://post-cyberlabs.github.io/Offensive-security-publicat...
Yes, the PIN is entangled with the key material. The admin PIN recovery techniques all involve enrolling additional unlocking methods.
Besides this, I completely agree with this post, especially:
> I think the best you can do right now is to layer a password with a hardware device. I don't think saying that the hardware devices are flawed means they are not useful as PART OF the security setup. It would certainly be nice if the software did this automatically/easily and it's unclear to me why it does not.
I've always found it silly that Microsoft's default BitLocker implementation doesn't make it easy to do the obvious thing that FileVault does and use the user's password as an additional wrapper to encrypt the KEK. Yes, then there's trouble when the user forgets their password and password-change rotation is a little annoying, but if you're in an enterprise scenario you need backup keys anyway, and your users are still as likely to get BitLocker'd by PCRs changing (ie due to EFI updates) than forgetting a password anyway.
No. Technically, TPMs can store secrets (limited due to little nvram), but no FDE implementation as far as I'm aware does this. TPMs wrap/encrypt the key and return an encrypted blob which will be stored in the header which will be decrypted upon request. The actual KEK is never stored on the TPM.
> I don't think Bitlocker does that
It does, which is why cryptsetup was much more affected by the faulTPM exploit: https://arxiv.org/abs/2304.14717
> Perhaps consider using a FIDO2 token (supported by cryptsetup) instead of a TPM. There are open-source implementations of FIDO2 and open-hardware ones too.
Not only do they do the same thing, they often use the same cheap chips. I don't see the benefit in using a separate device when you can have one SoC because there are too many downsides. Not only are dedicated devices vulnerable to sniffing and fault injection attacks, the extraction of secrets is actually feasible with the right kind of equipment. SoC solutions make this impossible (manufacturing levels are simply too small). Apple, Google, and Intel (Panther Lake) have solved this the right way. Open Source security can and should be shifted to PIN dependent key enrollment to protect against backdoored or flawed secure elements.
> 1. The strength of your disk encryption reduces to the strength of the password, where a TPM can have a 256-bit truly random key
I've used LUKS a fair amount, but don't have deep knowledge of how it works, but it's easy to add extra keys to an encrypted device (e.g cryptsetup luksAddKey) which suggests to me that the encryption isn't purely a function of the strength of the password.
Nadella gives a press release, "Alright guys, you got us fair and square. Backdoor on Bootlocker. Various versions of it for years on behalf of the spooks."
You are unlikely to ever get a confirmation of wrong doing. That being said, for a first line security posture, there is no way external media should have anything to do with the encryption process. Even if the OS chose to read a USB drive, to also delete the magical files is ridiculously suspect.
It could always be plain old incompetence, but that is a damning level of technical ineptitude assigned to such critical infrastructure. This is not a project you assign to the intern, but paranoid security experts. Multiple levels of code review and red-teaming.
Does this exploit have external media having anything to do with the encryption process? If yes, how do we know that? Remember that the OS normally unlocks the drive on boot, when no exploits are happening.
> Even if the OS chose to read a USB drive, to also delete the magical files is ridiculously suspect.
It's files in System Volume Information describing a transaction or something. It makes sense for it to resolve that transaction when mounting the external drive, and to then delete the files. And that's if it's even windows itself triggering the deletion.
This is why operating systems like GrapheneOS disable the USB port on the initial boot to limit the attack surface that an attacker has.
This is like finding out that an OS accepts an SSH private key circulating online that the sysadmin for those OS boxes never authorized, and saying "wait, we don't know that this is a backdoor into that system, the attackers just found a bug".
That is not what happens. There is nothing wrong with decrypting the drive. If you just powered on the computer normally, it will "trigger the decryption." There just isn't way to read a file from the lock screen. This exploit is getting you to a state where the drive is unlocked but the user has access to a command prompt. A command prompt, unlike a basic login screen gives the user the ability to actually see the contents of arbitrary files.
>specific file name
It's a specific file name because Windows stores transaction logs under that name. If it was a random name it wouldn't be able to exercise this vulnerable code.
>also removing said files after this is achieved
It doesn't seem farfetched for a transaction log to be deleted after it is successfully replayed.
Businesses use Microsoft because they figure if it’s backdoored it doesn’t matter and won’t affect them (because they aren’t terrorists or child pornographers or whatever, and they’d comply with a subpoena regardless of if Bitlocker is backdoored or not) and individuals who care about security and privacy put their shit on a Veracrypt drive somewhere else.
Which makes me think, it's becoming more and more urgent to make an open source mobile OS happen.
But Windows for personal private use? No.
(I'm aware that there's going to be a significant gap between the theory and what happens in practice though)
I do not know bitlocker/TPMs a lot, do they also prevent this sort of thing ?
Though if TrueCrypt was killed to try and get people to switch to encryption that could be backdoored, then why allow its successor VeraCrypt to exist? It's open source and independently audited, so it really shouldn't be backdoored.
I've seen every variant of:
1) "this is an authentication/privilege escalation bug, not a bitlocker exploit" (? what are you even trying to say)
2) "even though the attacker explicitly warns that this is capable of bypassing TPM+PIN, that isn't actually true or what he meant"
3) "we shouldn't jump to conclusions that this is a backdoor"
4) "we already knew BitLocker with just TPM isn't secure" (? except many organizations depend on it to be)
You just have to skip reading them because it seems there's no stopping those 100% genuine replies
2) Is it unreasonable to say "show it"?
3) Correct, we shouldn't jump to conclusions.
4) It's not known-insecure but it is known-enormous-attack-surface.
2) I'm sure many organizations are thankful that the researcher has decided not to release that exploit chain at this time. I am hopeful that Microsoft will not be as dismissive and will resolve it before it is publicly released.
3) It distracts from the point. The point is that Microsoft's security record is so bad that many of the vulnerabilities appear deliberate and obvious enough to be backdoors.
4) Yes, this also fits the definition of downplaying.
It's not semantics. A true bitlocker backdoor would let you in even if it's passworded.
And is it really downplaying? The ability to shove in a USB stick and get control over the drive is mostly equivalent to a bitlocker exploit when it comes to laptop theft. But for quick access to a desktop without bitlocker, and without the ability to open it and pull the drive, it's actually more damaging than a bitlocker exploit.
2) I am not personally being dismissive of the claim. I'm saying it's fine to hold off, and even if we assume the PIN version is real we shouldn't assume we know exactly what it looks like.
3) Saying it's not a backdoor distracts from the point? Can't agree with you there at all. The comments saying it's definitely a backdoor are the ones I point to as distracted.
4) Maybe it's downplaying but it's true. Replying on TPM-based bitlocker is a lot more dangerous than having a secure password. It's chosen because it's easier to enforce.
This exploit is only ever relevant with BitLocker enabled (as a method to "bypass" BitLocker's security premise [categorically classifying this as, dare I say, a "BitLocker bypass"]).
To avoid typing 1)2)3)4) a bunch of more times, I'll just say 2/3/4) all still fit the definition of downplaying the situation.
For this hypothetical, assume the owner took basic precautions to lock booting to the hard drive and password protect the BIOS.
But I'm not 100% familiar with how recovery mode normally works, so maybe it doesn't matter.
> To avoid typing 1)2)3)4) a bunch of more times, I'll just say 2/3/4) all still fit the definition of downplaying the situation.
I think that level of pushback against the claims is a valid (and small) amount of "downplaying". I haven't seen anyone claiming this isn't a serious issue.
> I think that level of pushback against the claims is a valid (and small) amount of "downplaying". I haven't seen anyone claiming this isn't a serious issue.
If you look in the other threads about this, it's much more obvious. Look for brand new users. There's comparatively few in this thread, but the pattern is there: if the user's name is green, they're downplaying this.
Just sign an alternate version of the recovery environment that doesn't bother displaying a login screen. Done - you can access any TPM-only Bitlocker setup freely. This is actually LESS risky than keeping the exploit in the publicly-available version of WinRE, because you don't have the risk of pesky security researchers finding your backdoor.
Hanlon's Razor and Occam's Razor both say this is probably a bug that lets you use some kind of early-boot filesystem-corruption-fixing code on the recovery image to break the login screen and leave the disk unlocked by accident. It deletes itself because it's, well, intended to be a filesystem fix log, and the log gets deleted when it's done being replayed so it doesn't happen a second time!
Just for fun (not saying I believe this!): Did you ever consider that a malicious Microsoft employee may have intentionally planted this exploit? They don't have access to signing keys. They aren't able to make custom firmware. However, what they can do is leave innocent looking code in Windows that they and their co-conspirators can exploit later. Completely possible (-:
For what it's worth, any OS using only hardware identity to unseal disk encryption is vulnerable to the same class of attack; there are all sorts of ways to misconfigure or exploit Linux FDE setups to enter a recovery shell as well. It's still a huge step above "no disk protection" (especially since it protects against every scenario where the disk is separated from the hardware), but the postboot surface area is enormous and nobody should be considering this class of protection as much more than a speed bump for a serious attacker.
Ideally you'd want that key to be further protected with a password or some other mechanism because it's not impossible to extract TPM keys.
I can't imagine there would be a way to bypass that if a password is required, unless it was a situation where like, there was originally some secret secondary key made that needs no password... or the password was never tied to the key in the first place.
[1]: https://deadeclipse666.blogspot.com/2026/05/were-doing-silen...
That's the thing, we don't actually know how involved the PIN is in relation to the key... it might be completely separate (and hence bypassable).
Similarly I also wonder if password-based pre-boot auth is affected.
https://blog.scrt.ch/2024/10/28/privilege-escalation-through...
Not saying you're wrong, I'm just not sure how well known it really is.
Either way... if the TPM is the one gating the key behind a PIN, I really don't see how an OS-level exploit can work without knowing the PIN in advance.
Do you know how hard discovering even one of those is? And how many months of work it takes?
Assuming reasonable implementation standards at this point is the irrational assumption, not the rational one.
You either need to enter a 128-bit entropy password on every boot (good luck with that) or you need to hold it on some external device, with some variant of USB / smartcard / NFC / Bluetooth to transmit it. NB. this is one of the cases where the usual "key for signing only, never leaves device, ephemeral DH and ZK protocols" like for SSH will not work on its own; you need the high-entropy key physically separate from the device.
The NSA realised this a while ago: https://en.wikipedia.org/wiki/KSD-64
Linux/LUKS etc. doesn't change any of this, by the way.
P.S. If Eclipse really has beef with Microsoft, he could always make an exploit that lets you set up a PC without making a Microsoft account.
That said, I think this is a thing with BitLocker? I remember coming across YubiKeys being able to do this via something called PIV (Personal Identity Verification). Found this guide now after giving it a quick search: https://gist.github.com/daemonhorn/03301a66da7d1f4de6cdc8c8b...
Not sure how sound of a design it is though, didn't dig into it much at all.
Yes, it's generally sound, and is the primary means of authentication and encryption used by the US military for classified systems.
https://deadeclipse666.blogspot.com/2026/05/were-doing-silen...