IBM AIX EUVD-2026-62862

The attacker must have layer-2 or adjacent-network reachability to the target AIX or VIOS system - AV:A precludes opportunistic internet-scale exploitation and requires the attacker to be on the same LAN, VLAN, or broadcast domain. No authentication is required (PR:N per the CVSS vector) and no user interaction is needed (UI:N). No specific non-default configuration flag or feature-enable condition is described in the CVE description, suggesting the vulnerable component is active in default deployments of IBM AIX 7.2/7.3 and VIOS 4.1. Exploitation is bounded to denial of service only - no confidentiality or integrity compromise is possible via this vector.

The CVSS base score of 4.3 (Medium) reflects an adjacent-network vector (AV:A) and low availability impact (A:L), which substantially limits real-world reach compared to internet-exposed flaws. However, the CVE description characterizes this as a 'remote attacker' scenario while the CVSS vector assigns AV:A - this discrepancy should be verified against the IBM advisory before finalizing network segmentation decisions, as the true attack distance changes defensive priorities. No EPSS score was provided, and neither a CISA KEV listing nor public proof-of-concept code has been identified, leaving exploitation probability unquantifiable. The CWE-787 class in an OS kernel or hypervisor layer can in practice exceed the rated A:L impact: a VIOS crash cascades I/O disruption to every LPAR on the physical host, meaning even a mid-sized enterprise running IBM POWER infrastructure in sanok would face broader service impact than the base score suggests if VIOS is the affected component. Overall this is a moderate, access-constrained risk with no confirmed active exploitation, but the hypervisor blast-radius warrants prioritized patching in environments where VIOS availability is business-critical.

An attacker positioned on the same network segment or VLAN as the IBM AIX or VIOS system sends a crafted network packet to the vulnerable service, triggering an out-of-bounds write that corrupts memory and causes the target system or critical service to crash, denying service to all dependent workloads. No proof-of-concept code has been identified at time of analysis, but CWE-787 is a well-understood exploitation class and researchers with protocol knowledge of the affected service could plausibly develop triggering payloads.