Topic

AI Infrastructure: news, advisories and analysis

7 articlesUpdated 23 Oct 2026

CyberSecureToday covers AI Infrastructure across 7 articles published between 20 Sep 2026 and 23 Oct 2026. The coverage includes 7 vulnerability advisories. 7 advisories are rated critical or high severity.

Stories on AI Infrastructure most often intersect with LiteLLM, SQL Injection and vLLM, with Technology, Cloud & SaaS the most affected sector. The most recent story, “LiteLLM Pipeline Hijacked in Software Supply Chain Attack Deploying InfoStealer and RAT”, was published 23 Oct 2026. Every article is listed below, newest first, with the affected products, severity and remediation guidance in each piece.

Latest coverage

ThreatsCloud & SaaSCritical
LiteLLM Pipeline Hijacked in Software Supply Chain Attack Deploying InfoStealer and RAT
Attackers hijacked LiteLLM release workflows using leaked GitHub CLI OAuth tokens, publishing backdoored packages to PyPI and Docker Hub executing LLMjacking.
Mei-Lin Chen23 Oct 202614 min read
NewsCloud & SaaSCritical
ChromaDB Vector Database Vulnerability Enables Unauthenticated Metadata Filter SQL Injection and Host Takeover (CVE-2026-78110)
ChromaDB patches a critical vulnerability (CVE-2026-78110) in vector metadata filtering that allows unauthenticated remote attackers to execute arbitrary code.
CST Editorial Desk12 Oct 20268 min read
ThreatsCloud & SaaSHigh
vLLM Distributed KV-Cache Memory Deserialization Flaw Enables Arbitrary Model Weight Extraction
A high-severity deserialization vulnerability in vLLM's distributed PagedAttention KV-cache transfer protocol allows attackers to compromise GPU worker nodes.
CST Editorial Desk10 Oct 20268 min read
ThreatsCloud & SaaSCritical
NVIDIA Triton Inference Server Memory Corruption Flaw Enables Remote Execution in GPU Clusters CVE-2026-34902
NVIDIA Product Security advisory addresses a critical memory corruption defect in Triton Inference Server enabling code execution across enterprise AI clusters.
Daniel Okafor6 Oct 20268 min read
ThreatsCloud & SaaSHigh
vLLM EngineCore Guided Decoding State Failure Triggers Multi-Tenant GPU Cluster Denial of Service (GHSA-85xf-c7hm-whqw)
A high-severity flaw in vLLM EngineCore guided decoding allows crafted JSON Schema requests to trigger non-recoverable state crashes across multi-tenant GPU workers.
Mei-Lin Chen30 Sep 202610 min read
ThreatsCloud & SaaSHigh
Ollama GGUF Parser Integer Overflow Exposes AI Inference Servers to Memory Corruption
A critical integer overflow flaw CVE-2026-86289 in the Ollama GGUF model decoder allows attackers to trigger memory corruption and remote code execution on AI inference clusters.
Priya Raman25 Sep 202616 min read
ThreatsCloud & SaaSCritical
LiteLLM Pre-Authentication SQL Injection Flaw Exposes Enterprise Master Keys and AI Model Routing Gateways
LiteLLM releases urgent patch for CVE-2026-42208, a pre-auth SQL injection flaw allowing unauthenticated remote attackers to dump LLM proxy master credentials.
CST Editorial Desk20 Sep 202614 min read

Latest coverage

LiteLLM Pipeline Hijacked in Software Supply Chain Attack Deploying InfoStealer and RAT

ChromaDB Vector Database Vulnerability Enables Unauthenticated Metadata Filter SQL Injection and Host Takeover (CVE-2026-78110)

vLLM Distributed KV-Cache Memory Deserialization Flaw Enables Arbitrary Model Weight Extraction

NVIDIA Triton Inference Server Memory Corruption Flaw Enables Remote Execution in GPU Clusters CVE-2026-34902

vLLM EngineCore Guided Decoding State Failure Triggers Multi-Tenant GPU Cluster Denial of Service (GHSA-85xf-c7hm-whqw)

Ollama GGUF Parser Integer Overflow Exposes AI Inference Servers to Memory Corruption

LiteLLM Pre-Authentication SQL Injection Flaw Exposes Enterprise Master Keys and AI Model Routing Gateways