Autodesk AutoCAD CVE-2026-16463

Exploitation requires the victim to open a specially crafted malicious DXF file in a vulnerable version of AutoCAD, AutoCAD LT, or DWG TrueView - the DXF import/parse path is the exact trigger. Limiting factors: the CVSS vector is AV:L with UI:R, meaning the attacker cannot exploit remotely or automatically; a local user must be socially engineered into opening the attacker-supplied file, and no unauthenticated network exploitation is possible. Exploitation runs only in the context of the current user (S:U, no privilege escalation), so impact is bounded by that user's permissions, and it depends on the user having a vulnerable, unpatched build installed.

The CVSS 3.1 base score is 7.8 (High) with vector AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H: exploitation is Local (the file must reach and be opened on the target machine), low complexity, no privileges, but requires user interaction, with high confidentiality, integrity, and availability impact. This is a file-parsing client-side bug, not a network-exposed service, so real-world risk depends on social engineering rather than automated worm-like spread. No EPSS score, KEV listing, or SSVC decision data was provided, and no public exploit is identified - signals that argue against imminent widespread exploitation. The primary risk driver is the ubiquity of DWG/DXF file exchange in engineering and construction workflows, where users routinely open files from third parties; the memory-disclosure primitive could also aid bypassing ASLR to reach reliable RCE. Overall this is a solid patch-soon priority for CAD-using organizations, not an emergency for the general population.

An attacker crafts a DXF file with a malformed length or record field that overflows a heap buffer in AutoCAD's parser, then emails it to a target engineer disguised as a project drawing or shares it via a design-collaboration channel. When the victim opens the file in AutoCAD, AutoCAD LT, or DWG TrueView, the overflow corrupts heap memory and executes the attacker's payload in the user's context; because AC:L and only user interaction is required, a single opened file is enough. No public exploit is identified at time of analysis, so this reflects the plausible attack path rather than an observed campaign.