Quick Answer: The FastUnlock disclosure, published on GitHub by researcher MSNightmare, claims an unpatched Windows Narrator flaw lets someone with physical access run code under a lock-screen account, even on BitLocker-encrypted machines. Microsoft hasn't confirmed it in the source material. Until it's verified, require a TPM plus startup PIN, shut down unattended laptops, and keep Windows updated.
A claim that BitLocker can be bypassed in five minutes should get your attention, whether or not it holds up. The FastUnlock Windows Narrator bypass write-up on GitHub says the flaw sits in how the accessibility tool draws its interface on a locked session. If the researcher is right, it breaks one of the assumptions most people rely on: that a properly encrypted Windows machine stays closed to someone holding it. Here's what the disclosure actually says, what it doesn't prove, and what you can do this week.
What the FastUnlock Disclosure Actually Claims
The GitHub repository describes a vulnerability in how Narrator handles its user interface in a locked-account context. The author says successful exploitation leads to code execution under a lock-screen account, and that it bypasses BitLocker. They tested it on Windows 11 with the November update installed. They also say it worked consistently across the machines they tried, that it took about five minutes, and that it requires physical access plus at least one account linked to an online Microsoft account.
Three points matter before you panic.
First, none of this is confirmed. The source is one researcher's self-reported findings. It includes no CVE, no Microsoft advisory, and no independent reproduction. Second, the author says they haven't tested Windows 10, so the scope is unknown. Third, the author compares it with earlier lock-screen bypasses they reported, including ZDI-21-1053, and says this one is more convenient to use. Convenience is a reason to take the report seriously. It doesn't prove the claim.
Most coverage will frame this as "BitLocker broken." That's the wrong frame, and the reason is the most useful thing in this article.
Why BitLocker Doesn't Cover What Happens After Boot
BitLocker protects data at rest. Its main job is to make a powered-off drive unreadable when someone pulls it out and plugs it into another machine. Once Windows has booted and the volume is unlocked, the encryption has already done its job for that session. The disk is readable by the running system.
That's why a flaw on the lock screen matters even on an encrypted drive. The lock screen runs after the disk is decrypted. A bug there doesn't defeat the encryption math. It gives someone a foothold inside a system whose drive is already open.
This is also why the common advice "turn on BitLocker and you're safe" is incomplete. In TPM-only mode, the machine boots without any input from you. Anyone who powers on the laptop reaches the sign-in screen with the volume already unlocked. A startup PIN changes that by requiring a secret before Windows loads, but it does nothing once you're past the boot prompt.
Here's the trade-off, and it's a real one. Adding pre-boot authentication makes the machine harder to use: someone has to type a PIN on every cold start. For most people with a laptop that lives in a bag, that's a fair price. For a shared desktop in a reception area, it may be a dealbreaker.
Why Sign-In Screen Accessibility Tools Are a Recurring Attack Surface
Accessibility features are unusual because they have to run before anyone signs in. A person who can't see the screen or use a keyboard normally needs Narrator, Magnifier, or on-screen keyboards available at the sign-in screen. That requirement is a legitimate design constraint, and it's also why these components keep showing up in security research.
Security researchers have documented this pattern for years. Attackers have replaced or manipulated accessibility executables on disk, and they've used the sign-in screen's built-in tools as launch points for code. The general lesson is that any feature reachable from the lock screen is part of the pre-login attack surface, and it deserves the same scrutiny as the login prompt itself.
The trade-off is hard to avoid. Removing or restricting accessibility tools protects against this class of issue, but it can lock out people who depend on them. Most organisations should not respond by disabling Narrator for everyone. They should understand their exposure and decide based on who actually uses those features.
Here's where most guides go wrong: they treat "physical access" as a minor detail. It's not. Physical access is the entire threat model here, and it changes which controls matter.
Choosing a BitLocker Setup: Comparing the Options
The table below compares common setups by what they protect against and where they fall short. Use it to see which gaps matter for your situation.
| Setup | Protects against | Gap it leaves |
|---|---|---|
| BitLocker, TPM only | Someone removing a powered-off drive | Machine boots with no user input, so anyone who powers it on reaches a decrypted system |
| BitLocker, TPM plus startup PIN | Drive removal and boot-time tampering | Does nothing about a running, unlocked session or flaws on the sign-in screen |
| Auto-lock with a short idle timeout | Casual access to an unattended, logged-in session | Depends on the lock screen behaving correctly, which is the exact point under dispute |
| Shut down instead of sleep when unattended | Keeps the session closed and reduces key material held in memory | Slower restart, and some users will skip it out of habit |
No single row is enough. The strongest posture for a laptop that leaves your desk combines the first two rows with the last one.
Hardening Steps You Can Apply Now
Until Microsoft confirms or rejects the claim, these steps reduce your exposure whether or not the disclosure holds up. Each one has a rationale, so you can decide what fits your environment.
-
Turn on BitLocker with TPM plus a startup PIN. Go to Settings, then Privacy and security, then Device encryption, or use the BitLocker control panel. Choose a PIN you'll remember and store the recovery key in your Microsoft account or an offline location. This adds a secret that must be entered before Windows loads, which blocks the simplest way to reach a decrypted system.
-
Shut down unattended devices. Don't leave a laptop asleep on a desk or in a shared space. A shutdown clears the running session and forces pre-boot authentication on the next start. It's the simplest control on this list.
-
Set a short idle lock timeout. Use a screen saver with the "On resume, display logon screen" option, or Dynamic Lock with a paired Bluetooth device. The goal is to shrink the window when a logged-in session is open and unattended.
-
Keep Windows fully updated and watch for advisories. Install cumulative updates promptly and check the Microsoft Security Response Center (MSRC) for advisories. Confirmed vulnerabilities get fixes and CVE identifiers. Self-reported claims without those details should be treated as open questions until they're resolved.
-
Limit who can reach the device physically. This sounds obvious, but it's the control that matters most for this threat model. A locked office matters more than any setting.
What to Watch for and How to Verify
The FastUnlock write-up is one researcher's account, and that's enough to make it worth tracking. It's not enough to act on as if it were proven. Watch for three signals: a Microsoft acknowledgement or advisory, a CVE assignment, and independent reproduction by other researchers. Any of those would change the risk picture. Their absence means you're working from a claim.
The write-up includes a detailed sequence of steps for reproducing the issue. This article doesn't reproduce them. Testing this against machines you don't own is illegal in most jurisdictions, and the hardening steps above reduce your exposure whether or not you verify the claim yourself.
Most people stop here, and that's a mistake. Ask yourself what you'd do if the claim were true tomorrow. If your answer is "I'd shut down my laptop and enable a PIN," you've already done most of the work.
Frequently Asked Questions
What is the FastUnlock Windows Narrator bypass?
It's a self-reported vulnerability published on GitHub by researcher MSNightmare. The author says a flaw in Narrator's interface on the lock screen allows code execution under a lock-screen account, even on BitLocker-encrypted devices, with physical access. Microsoft has not confirmed it in the source material, and there's no CVE listed.
How can I protect BitLocker from physical access attacks?
Enable BitLocker with TPM plus a startup PIN so a secret is required before Windows loads. Shut down devices you leave unattended instead of sleeping them, and set a short idle lock timeout. These steps reduce exposure to someone with physical access, though they can't block every lock-screen flaw.
Why does Windows Narrator run on the lock screen?
Accessibility tools must work before anyone signs in, so users who depend on screen readers can reach the sign-in process. That requirement makes them part of the pre-login attack surface. Balancing that access against security is a real trade-off.
Should I turn off Narrator to stay safe?
Not without knowing whether you need it. Disabling Narrator may reduce exposure to this kind of issue, but it also blocks users who rely on it. For most people, the better move is layered hardening: a startup PIN, shutdown when unattended, and prompt updates.
Does this affect Windows 10 too?
The author says they haven't tested Windows 10, so it's unknown. Don't assume it's unaffected. Check Microsoft's advisories for your specific build and keep it updated.
Turn on a startup PIN for BitLocker on one device this week, reboot it, and note whether it prompts before Windows loads. If you're protecting a fleet or a shared workstation, pass this to whoever manages your device encryption settings, because a BitLocker setup you haven't checked is one you can't vouch for.