Windows Server EUVD-2026-73366

Exploitation requires a local authenticated session on the target Windows system with at minimum a standard (low-privilege) user account - consistent with CVSS PR:L. No special configuration, non-default features, or elevated pre-existing rights are needed beyond the ability to execute local code. The attack vector is strictly local (AV:L); remote unauthenticated exploitation is not indicated. What limits real-world risk: the attacker must already have local code execution (e.g., via malware, RDP login, or physical access), making this a post-exploitation rather than initial-access vulnerability. No user interaction from a victim is required once local access is established.

The CVSS 7.8 score (AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H) accurately reflects a reliable local privilege escalation: low-privilege access is sufficient, complexity is low, and no user interaction is needed, yielding full confidentiality, integrity, and availability impact post-exploitation. In practice, this class of vulnerability is a standard toolkit component for attackers who have already achieved initial access (phishing, credential theft, RDP brute-force) and need to escalate to SYSTEM. The risk is highest in shared-access environments: RDP farms, terminal servers, cloud VMs accessible to multiple tenants, and developer workstations with domain trust. No EPSS score was provided in the input data; without KEV confirmation or a public PoC, exploitation probability in the near term is likely moderate but not imminent. The 'Denial Of Service' tag in the input metadata is inconsistent with the description and CVSS - the primary impact is privilege escalation, not service disruption, though a UAF can incidentally crash a process. This discrepancy should be noted but does not change the overall risk picture.

An attacker who has obtained a standard (low-privilege) local user account - via phishing, credential stuffing, or initial-access malware - executes a crafted application that triggers the use-after-free condition in the vulnerable Windows component, corrupting freed kernel memory to redirect execution flow and obtain SYSTEM-level privileges. From SYSTEM, the attacker can disable endpoint defenses, dump credential hashes from LSASS, or move laterally across the domain. No public proof-of-concept has been identified at time of analysis, but the low attack complexity (AC:L) indicates exploitation is straightforward once the trigger is identified.