What changed in AI security, Nov 9 to Nov 15, 2026
Nov 9 to Nov 15, 2026 (ISO week 2026-W46). Weeks run Monday to Sunday in UTC.
260 records published, -78 on the previous week: 34 vulnerabilities (-15), 0 incidents (no change), 59 research items (-58), 166 news items (-6), 1 policy item (+1).
Critical and high advisories
Vulnerability records rated critical or high, newest first.- High
CVE-2026-126397: vLLM denial of service via unbounded max_patches in multimodal preprocessing
CVE-2026-126397NVD/CVE Database - High
ZDI-26-976: llama.cpp GGUF File Parsing Heap-based Buffer Overflow Remote Code Execution Vulnerability
ZDI Published Advisories - High
ZDI-26-975: llama.cpp GGUF File Parsing Integer Overflow Remote Code Execution Vulnerability
ZDI Published Advisories - High
ZDI-26-954: llama.cpp MTMD Multimodal GGUF File Parsing Integer Overflow Remote Code Execution Vulnerability
ZDI Published Advisories - High
ZDI-26-968: (Pwn2Own) OpenAI Codex GUI OpenProjectInCodex Argument Injection Remote Code Execution Vulnerability
ZDI Published Advisories - High
CVE-2026-67548: taOS system settings routes missing admin authorization check
CVE-2026-67548NVD/CVE Database - High
CVE-2026-67547: taOS remote code execution via skill-exec endpoint for non-admin users
CVE-2026-67547NVD/CVE Database - High
CVE-2026-65820: aeon GitHub Actions workflow exposes tokens to fork pull requests
CVE-2026-65820NVD/CVE Database - High
CVE-2026-63651: LiveDocs is a primary-source-first documentation for AI agents. Prior to version 0.7.0, fetch_docs and introspect…
CVE-2026-63651NVD/CVE Database - High
CVE-2026-125271 - AWS API MCP Server Security Policy Bypass via Incorrect Operation Name Normalization
AWS Security Bulletins - High
GHSA-ghwv-vrp2-2975: Cocos AI Intra-handshake attested TLS implementation is vulnerable to TOCTOU Attacks
CVE-2026-119345GitHub Advisory Database - High
CVE-2026-123818: LMDeploy is a toolkit for compressing, deploying, and serving large language models. From 0.12.3 until 0.15.0…
CVE-2026-123818NVD/CVE Database - High
GHSA-vvr2-h2jp-838m: Code Graph RAG: MCP paginated read_file path traversal / arbitrary file read
CVE-2026-77604GitHub Advisory Database - High
GHSA-4p8p-pjpq-2w7h: tsParticles: Malformed JSON requests can terminate the @tsparticles/mcp-server HTTP process
CVE-2026-76892GitHub Advisory Database - High
GHSA-447x-qqwf-m7xq: tsParticles: Unbounded HTTP Request Body Buffering in @tsparticles/mcp-server Leads to Remote DoS
CVE-2026-63655GitHub Advisory Database - Critical
CVE-2026-102895: GPUStack Improper Authentication Vulnerability
CVE-2026-102895CISA Known Exploited Vulnerabilities
Exploitation signals
Vulnerabilities published in the week that are listed in the CISA Known Exploited Vulnerabilities catalog or have an EPSS score of 10% or more.| Advisory | Exploitation | EPSS | Published |
|---|---|---|---|
| CVE-2026-102895: GPUStack Improper Authentication Vulnerability CVE-2026-102895CISA Known Exploited Vulnerabilities | Known exploited | 0.5% |
Packages that began delegating to a language model
Exposure Registry packages whose first release declaring an LLM SDK, agent framework or MCP dependency was published in the week.No tracked package published its first release with an LLM SDK, agent framework or MCP dependency in this week.
Topics that moved
Largest increases over the mean of the 4 previous weeks, for topics with at least 3 records in the week.| Topic | Records | Weekly mean, previous 4 | Difference |
|---|---|---|---|
| Inference infrastructure | 10 | 7.0 | +3.0 |
Research
Peer-reviewed first, then newest. Showing 8 of 59.A reliability enhancement framework for LLM-based attack graph generation in software supply chain security
Peer-reviewedElsevier Security JournalsInvestigating differentially private learning for document key information extraction
Peer-reviewedOpenAlex (peer-reviewed AI security)A Systematic Review of Privacy-Preserving Approaches in Deep Learning Applications
Peer-reviewedOpenAlex (peer-reviewed AI security)<em>SEAL</em>: An integral methodology for evaluating the security of agentic LLMs
Peer-reviewedElsevier Security JournalsImage watermarking robustness under MLLM-guided semantic editing: Benchmark, defense, and diagnosis
Peer-reviewedElsevier Security JournalsSecurity, Privacy, and Trustworthy Design for Retrieval-Augmented Generation Systems
Peer-reviewedOpenAlex (peer-reviewed AI security)Adversarial Retraining for Machine Learning-Based Man-in-the-Middle Attack Detection in Smart Grid Industrial Control Systems
Peer-reviewedOpenAlex (peer-reviewed AI security)Cracks in Embeddings: Robustness and Stability of JEPA-Based Chest X-Ray Representations
PreprintArxiv (cs.RO + cs.CV security)
Policy and regulation
Newest first.Generated from the AI Sec Watch database at . Every item links to its record.