CVE Vulnerability Catalog
Translated CVE descriptions from NVD NIST - in English
CISA KEV catalog updated: (v2026.08.25)
Weekly CVE digest
One email a week with newly published vulnerabilities worth knowing about. No account needed.
This digest covers new vulnerabilities in general, not your servers. If you want to know which of them actually run in your infrastructure, that is what Secvalis does: it scans your machines and reports only what concerns them.
Deserialization of untrusted data for some Intel(R) Extension for PyTorch before version 2.8.0 may allow an escalation of privilege. An unprivileged adversary with an unauthenticated user and low complexity attack may enable privilege escalation. This may occur via local access and requires active user interaction. Impact on confidentiality, integrity, and availability is low.
Uncontrolled search path for some Hardware-Aware-Automated-MachineLearning before version 45cd723 may allow an escalation of privilege. An unprivileged adversary with a privileged user and low complexity attack may enable privilege escalation. This may occur via local access and requires passive user interaction. Impact on confidentiality, integrity, and availability is high.
Untrusted search path for some Intel(R) Performance Counter Monitor (Intel(R) PCM) before version tag 202604 may allow an escalation of privilege. An unprivileged adversary with an authenticated user and high complexity attack may enable privilege escalation. This may occur via local access and requires passive user interaction. Impact on confidentiality, integrity, and availability is high.
Uncontrolled search path for some Approximate Bayesian Inference Framework before version on commit #484c949 may allow an escalation of privilege. An unprivileged adversary with a privileged user and low complexity attack may enable privilege escalation. This may occur via local access and requires passive user interaction. Impact on confidentiality, integrity, and availability is high.
Path traversal vulnerability in gaudi-container-runtime before version 1.24.0 may allow privilege escalation. An attacker with an authenticated user and a low complexity attack can exploit this locally, requiring passive user interaction. The impact on confidentiality, integrity, and availability is high.
Protection mechanism failure in Intel(R) Workload Services Framework may allow privilege escalation. An attacker with a privileged user and a low complexity attack can exploit this locally, requiring passive user interaction. The impact on confidentiality, integrity, and availability is high.
Protection mechanism failure in LLM-on-Ray before version 1.0 may allow privilege escalation. An attacker with a privileged user and a low complexity attack can exploit this locally, requiring passive user interaction. The impact on confidentiality, integrity, and availability is high.
Uncontrolled search path in EquiTriton before version f5ddbb5 may allow privilege escalation. An attacker with a privileged user and a low complexity attack can exploit this locally, requiring passive user interaction. The impact on confidentiality, integrity, and availability is high.
Improper input validation in vLLM Hardware Plugin for Intel(R) Gaudi(R) before version 0.16.0 may allow denial of service. An authorized attacker with an authenticated user and a low complexity attack can exploit this locally, without user interaction. The impact on availability is high.
Protection mechanism failure in Intel(R) oneCCL Bindings for PyTorch before version v2.8.0 may allow privilege escalation. An attacker with a privileged user and a low complexity attack can exploit this locally, requiring passive user interaction. The impact on confidentiality, integrity, and availability is high.
Use after free in Intel(R) PROSet/Wireless WiFi Software for Windows may allow denial of service. An attacker with an unauthenticated user and a high complexity attack can exploit this locally, without user interaction. The impact on availability is high.
Protection mechanism failure in Intel(R) AI Reference Models before version v3.4.1 may allow privilege escalation. An attacker with a privileged user and a low complexity attack can exploit this locally, requiring passive user interaction. The impact on confidentiality, integrity, and availability is high.
Heap-based buffer overflow in Intel(R) Open Volume Kernel Library (Open VKL) before version 2.0.2 may allow denial of service. An attacker with an authenticated user and a low complexity attack can exploit this locally, without user interaction. The impact on integrity is low and on availability is high.
Protection mechanism failure in Intel(R) LLM Library for PyTorch may allow privilege escalation. An attacker with a privileged user and a low complexity attack can exploit this locally, requiring passive user interaction. The impact on confidentiality, integrity, and availability is high.
A stored Cross-Site Scripting (XSS) vulnerability in the software that could be exploited by a low-privileged attacker to inject malicious scripts into vulnerable form fields. Malicious JavaScript may execute in a victim's browser when they browse to the page containing the vulnerable field. Scope is changed.
Improper input validation in Intel(R) Neural Compressor software before version v3.7 may allow an escalation of privilege. An unprivileged attacker with an authenticated user and a low complexity attack may enable privilege escalation. The attack requires local access, no user interaction, and may impact confidentiality, integrity, and availability.
A Time-of-check time-of-use race condition in the Intel(R) NPU Driver for Windows may allow a denial of service. An attacker with an authenticated user and a high complexity attack may cause a denial of service. The attack requires local access, no user interaction, and may impact integrity and availability.
Protection mechanism failure in Intel(R) Neural Compressor software before version v3.6 may allow an escalation of privilege. An attacker with a privileged user and a low complexity attack may enable privilege escalation. The attack requires local access, passive user interaction, and may impact confidentiality, integrity, and availability.
Protection mechanism failure in Intel(R) AI Containers before version v0.4.0 may allow an escalation of privilege. An attacker with a privileged user and a low complexity attack may enable privilege escalation. The attack requires local access, passive user interaction, and may impact confidentiality, integrity, and availability.
Improper authentication in Intel(R) PROSet/Wireless WiFi Software for Windows may allow an escalation of privilege. An unprivileged attacker with an unauthenticated user and a low complexity attack may enable local code execution. The attack requires local access, no user interaction, and may impact confidentiality, integrity, and availability.

