New Hello Again Attack Bypasses Windows Hello and BitLocker PIN Lockout Protection

Blog WriterCybersecurity News - Original News Source is cybersecuritynews.com


Researchers have shown that someone with a laptop and a few cheap parts can guess a Windows Hello PIN in about 40 hours. With the lockout working as designed, the same job would take 69 days.

The same trick also cuts the worst-case time to crack a default 6-digit BitLocker startup PIN from 19 years to 20 days. The work is called “Hello Again” and comes from a paper on arXiv that is also tied to ACM CCS ’26.

Windows Hello is the main sign-in method on more than one billion Windows 11 devices, and it allows PINs as short as four digits. A short PIN is only safe because the TPM, the security chip that guards keys, locks down after too many wrong guesses. Microsoft explains that the TPM protects against PIN brute-force attempts on lost or stolen devices, and the device locks after too many bad tries.

BitLocker works the same way. Microsoft’s BitLocker countermeasures page says TPM anti-hammering is built to stop people from guessing a startup PIN. The whole design depends on the chip keeping an honest count of failed guesses.

Hello Attack Bypasses Windows Hello and BitLocker

Most modern x86 PCs use a firmware TPM, or fTPM, which runs inside the processor and has no private storage of its own. It saves its state, including the failed-guess count, on the SPI flash chip on the motherboard. If someone copies that saved state and writes it back later, the count goes back too. To stop this, newer boards use Replay Protected Monotonic Counters, or RPMC.

These are counters inside the flash chip that can only move up. The researchers did the first practical security study of RPMC in real x86 PCs and found several weak spots.

The first flaw is in Winbond flash chips. Winbond is the top NOR flash maker, with more than 25 percent of the market. The team found that a short voltage glitch makes the chip leak its HMAC key, the secret used to sign counter values. With that key in hand, they could forge any counter value they wanted.

The second flaw is in AMD’s non-volatile controller. It never checks that a counter actually went up. A small board sitting between the processor and the flash chip can block the increase, say it worked, and the PC accepts the old state without noticing.

AMD PCs without RPMC flash looked even weaker. On two AMD desktops, writing an old flash image back reset the fTPM’s lockout counters, and no special board was needed. The paper says these rollbacks work on AMD, Intel, and Pluton fTPM state.

The researchers tested the idea on locked and encrypted hardware from start to finish. Each time the TPM locked them out, they rolled back the saved state and kept guessing. This brought a 4-digit Hello PIN down to 40 hours, and the BitLocker TPM+PIN protector fell the same way.

Attack Setup

A stolen Hello PIN also opens device-bound keys and Hello-based single sign-on at work, so the damage can reach company accounts and not only the laptop.

AMD has published bulletin AMD-SN-7079 for the flaw. It tracks the issue as CVE-2026-43613 and gives it a CVSS score of 5.1. AMD says the Secure Processor firmware does not properly confirm that an RPMC increase finished, allowing someone with physical access to suppress counter updates.

AMD notes that physical attacks sit outside its normal threat model. It adds that RPMC exists to stop exactly this kind of interposer trick, so it plans firmware fixes. Some target dates for affected Ryzen, EPYC, and Athlon lines are in November 2026.

Until the fixes arrive, AMD tells users to keep systems under physical control and watch for signs of tampering. The Winbond key leak is a chip design problem, and the paper’s findings suggest that only new flash designs can close it.

Other BitLocker Bypasses

This is not the first time BitLocker’s physical protections have been tested this year. Cyber Security News covered the YellowKey BitLocker zero-day, which let someone with physical access skip BitLocker through the Windows Recovery Environment. We also reported on Microsoft’s YellowKey mitigation, tracked as CVE-2026-45585.

Microsoft’s advice after YellowKey was to move from TPM-only mode to TPM+PIN. Hello Again shows why that advice needs a second step. A short PIN is only as strong as the lockout behind it. The GreatXML bug, which we covered in our GreatXML report, points to the same pattern of physical-access bypasses.

The risk is real but narrow. It requires physical access, special tools, and time, so it mainly worries owners of lost or stolen laptops that carry valuable data. Longer PINs are the easiest fix, because each added digit multiplies the number of guesses needed.

Microsoft’s guidance also suggests using a longer, enhanced alphanumeric PIN with BitLocker when strong protection is needed. Admins can raise the minimum PIN length in Group Policy, turn on enhanced startup PINs, and install AMD firmware updates once vendors release them. Companies that store sensitive data should also treat a stolen laptop as a possible data exposure, even when the drive was encrypted.

What Are Leading Organizations Doing About AI and Cyber Risks? Find Out. View Details!

The post New Hello Again Attack Bypasses Windows Hello and BitLocker PIN Lockout Protection appeared first on Cyber Security News.