Last updated: 9 November 2026
CVE-2026-109469 is a critical, maximum-impact vulnerability in Microsoft Entra Device Registration. It carries a CVSS 3.1 base score of 10.0 and is classed as an authorisation bypass through a user-controlled key (CWE-639), meaning an attacker could potentially elevate their privileges over the network with no authentication and no user interaction required. Microsoft publicly disclosed the flaw on 5 November 2026 and has already fixed it entirely on its own servers. Entra Device Registration is a Microsoft-hosted cloud service, so no customer patching is needed. Here's everything we know so far, and what's still unconfirmed.
In plain terms: this let a device trick Microsoft's trust system into treating it as something it wasn't. It happened entirely on Microsoft's side, so there's nothing for you to install. It's still worth a quick glance below if your organisation runs Microsoft 365 or Entra ID.
| CVSS 3.1 score | 10.0 (Critical) |
| Vector | AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H |
| Weakness | CWE-639: Authorization Bypass Through User-Controlled Key |
| Affected product | Microsoft Entra Device Registration (cloud service) |
| NVD status | Awaiting Analysis (as of 9 Nov 2026) |
| Exploited in the wild? | Not confirmed; not listed in CISA's KEV catalogue |
| Source / CNA | Microsoft Security Response Center (advisory) |
All of the above comes from the official NVD record and MITRE CVE record. Both agree on the description, score and affected product.
In plain English: anyone on the internet could have triggered this, with no password and no clicks needed from a victim. A successful attack would have handed over full control, which is why it maxes out the scoring scale.
In everyday terms: this bug could have let an attacker's device pretend to be one it had no right to be, tricking Microsoft's system into trusting it when it shouldn't have. Technically, Microsoft classifies it as CWE-639, "Authorization Bypass Through User-Controlled Key." That's a flaw where a system uses an identifier or key supplied by the caller, rather than one it independently verifies, to decide what that caller is allowed to access or do. In the context of Entra Device Registration, the service that lets a device enrol itself into an organisation's Entra ID tenant, that typically means an attacker could supply or manipulate a device or registration identifier to act with permissions it shouldn't have.
Microsoft hasn't published the exact technical root cause, and no credible proof-of-concept has surfaced elsewhere. Most of what's circulating online just repeats the official description. The short version: this was exploitable by anyone on the internet, with no login details and no action needed from a victim, which is why it scored the maximum possible severity.
| Date | Event |
|---|---|
| 12 Oct 2026 | CVE ID reserved by Microsoft |
| 5 Nov 2026 | Publicly disclosed by Microsoft (MSRC advisory published) |
| 6 Nov 2026 | Published to the MITRE CVE and NVD databases |
| 9 Nov 2026 | This post. NVD analysis is still pending, and no exploitation has been confirmed |
A couple of third-party trackers list a 4 November release date instead; we've gone with the dates in the official MSRC, MITRE and NVD records above.
As of 9 November 2026, CVE-2026-109469 is not listed in CISA's Known Exploited Vulnerabilities (KEV) catalogue, and we haven't seen any credible reports of it being exploited in the wild.
More importantly for anyone running Microsoft Entra ID: NVD tags this record "exclusively-hosted-service." That's a technical way of saying Entra Device Registration runs entirely on Microsoft's infrastructure, not in your tenant's own software stack. There's no patch, update or KB article for you to deploy. Microsoft fixed the vulnerability server-side, and the fix applies automatically to every tenant. There is nothing to install.
No patching is available, but given the severity score, it's worth a quick sense check:
None of this is a required remediation step. It's a sensible 15-minute check given the severity score, not a response to any confirmed compromise.
Entra ID (formerly Azure AD) sits underneath almost everything else in a modern Microsoft environment. It's what decides who and what gets trusted, including the devices that feed into Conditional Access and zero-trust policies. A maximum-severity flaw in its device-trust layer is worth paying attention to even when it's fixed within days, because identity infrastructure is exactly the kind of system where a small authorisation bug can have an outsized blast radius.
For organisations working towards NHS DSPT, Defence Cyber Certification (DCC), or Cyber Essentials, identity and access management is routinely in scope for assurance activity. This is precisely the kind of advisory worth feeding into your next security review, internal audit, or penetration test scoping conversation. If you're not sure whether your Entra environment would hold up to that kind of scrutiny, that's a conversation worth having before an assessor, or an attacker, asks the question for you.
Sources: MSRC advisory · NVD record · MITRE CVE record
Update (9 November 2026): We've linked Microsoft's official advisory above because it's the canonical reference both NVD and MITRE cite for this CVE. That advisory page requires JavaScript and didn't render for our research tools, so the facts in this post are drawn from the NVD and MITRE CVE databases, which independently agree on the description, score and affected product.