BitLocker Downgrade Attack on Windows 11 Exposes Encrypted Disks

Key Takeaways

• A new tool, BitUnlocker, enables a critical downgrade attack against BitLocker encryption on Windows 11.
• Attackers with physical access can decrypt BitLocker volumes in under five minutes, even on patched systems.
• The vulnerability (CVE-2025-48804) exploits an unrevoked legacy signing certificate for the boot manager, not a missing patch.
• Systems using TPM-only BitLocker and trusting the old certificate are vulnerable; TPM + PIN or migration to CA 2023 provides protection.

BitLocker Downgrade Attack Bypasses Windows 11 Encryption

A new exploit tool, dubbed “BitUnlocker,” has been revealed, demonstrating a practical downgrade attack against Microsoft’s BitLocker full-disk encryption. This technique allows threat actors with physical access to a target Windows 11 machine to decrypt protected volumes in less than five minutes, even if the system has the latest patches installed. The core issue stems from a critical window between the deployment of security updates and the revocation of legacy signing certificates.

The attack leverages CVE-2025-48804, one of four critical zero-day vulnerabilities discovered by Microsoft’s Security Testing & Offensive Research (STORM) team. These flaws were addressed during the July 2025 Patch Tuesday cycle. However, the fix alone does not fully mitigate the risk.

Exploiting the Windows Recovery Environment

According to research from Intrinsec, the vulnerability resides within the Windows Recovery Environment (WinRE) and specifically targets the System Deployment Image (SDI) file mechanism. The process exploits how the boot manager handles WIM (Windows Imaging Format) files referenced by an SDI for integrity verification.

During the boot sequence, while the boot manager validates a legitimate WIM file, it inadvertently permits an attacker-controlled WIM to be appended to the SDI’s blob table. Critically, the system verifies the integrity of the first, legitimate WIM but then proceeds to boot from the second, malicious WIM. This attacker-supplied WIM contains a modified WinRE image configured to launch cmd.exe with the BitLocker-encrypted volume already decrypted and mounted.

Microsoft released a patched bootmgfw.efi binary for all supported Windows versions via Windows Update in July 2025. Despite this, the patch itself does not fully close the attack vector.

The Critical Flaw: Unrevoked Certificates

The primary weakness enabling the BitUnlocker attack is not a lack of a patch, but rather the continued trust in an unrevoked signing certificate. Secure Boot mechanisms validate a binary based on its signing certificate, not its version number. The legacy Microsoft Windows PCA 2011 certificate, which was used to sign all boot managers prior to the July 2025 fix, remains trusted in the Secure Boot databases of most active machines. This trust persists unless a clean Windows installation was performed after early 2026.

This means that an older, pre-patch bootmgfw.efi, which is signed with the PCA 2011 certificate, is still considered valid by Secure Boot despite containing the vulnerability. A mass revocation of the PCA 2011 certificate presents a significant operational challenge for Microsoft, as it would invalidate a vast number of legitimate signed binaries across the entire Windows ecosystem.

Chaining Vulnerabilities into a Five-Minute Attack

Building upon the original STORM research and previous work on the “bitpixie” downgrade exploit, researchers developed a proof-of-concept (PoC) that chains these weaknesses into an attack achievable in under five minutes. Intrinsec’s findings indicate that an attacker requires only physical access to the target workstation, a USB drive or a PXE boot server, and no specialized hardware.

The attack unfolds as follows: the attacker prepares a modified Boot Configuration Data (BCD) file that points to a tampered SDI. They then serve an older, vulnerable boot manager, signed by the PCA 2011 certificate, via a USB drive or PXE boot. The target machine loads this pre-patch boot manager, which successfully passes Secure Boot validation due to the trusted legacy certificate.

Crucially, the Trusted Platform Module (TPM) releases the BitLocker Volume Master Key (VMK) without triggering any alerts. This is because the Platform Configuration Register (PCR) measurements 7 and 11 remain valid under the trusted PCA 2011 certificate. The outcome is a command prompt that opens with the operating system volume fully decrypted and mounted.

Systems configured with TPM-only BitLocker (without a PIN) are fully vulnerable if their Secure Boot database still trusts the PCA 2011 certificate. However, machines configured with TPM + PIN are protected, as the TPM will not unseal the VMK without user interaction during pre-boot authentication. Furthermore, systems that have completed the KB5025885 migration, which transitions the boot manager signature to the newer Windows UEFI CA 2023 certificate, are also protected against this specific downgrade path.

What You Should Do

Given the public availability of the PoC on GitHub, enterprise defenders must urgently audit their BitLocker configurations and accelerate migration efforts. Immediate actions include:

• Enable TPM + PIN Pre-Boot Authentication: This is the most effective control, preventing the TPM from releasing the VMK during any manipulated boot sequence without user input.
• Deploy KB5025885: This critical Microsoft update migrates the boot manager signing to the CA 2023 certificate and introduces revocation controls that eliminate this specific downgrade path.
• Verify Boot Manager Certificate: Mount the EFI partition and use a utility like sigcheck to confirm that the active bootmgfw.efi is signed under the newer CA 2023, not the legacy PCA 2011.
• Remove the WinRE Recovery Partition: For high-security workloads where pre-boot authentication cannot be enforced, consider removing the WinRE recovery partition to minimize the attack surface exposed to this class of exploit.

Disclaimer: HackersRadar reports on cybersecurity threats and incidents for informational and awareness purposes only. We do not engage in hacking activities, data exfiltration, or the hosting or distribution of stolen or leaked information. All content is based on publicly available sources.