18006 matches found
Security Vulnerability Patterns in AI-Generated Code: A Cross-Model Comparative Study
LLM-based coding tools enable non-expert users to generate routine automation scripts that may enter enterprise workflows without meaningful security review. This study examines that risk directly. Code was collected from ChatGPT, Microsoft Copilot, and Google Gemini using identical prompts acros...
HijackKV: New Threat in Position-Independent KV Cache Reuse
Key-Value KV cache reduces inference latency in large language models LLMs. Traditional prefix-based reuse has low cache hit rates across inference requests because it requires exact token and position matches. To improve efficiency, recent system optimizations introduce position-independent KV...
PT-2026-63437
InvokeAI before 6.13.7 contains an unauthenticated directory enumeration vulnerability in the GET /api/v2/models/scan folder endpoint that accepts attacker-controlled scan path parameters. Unauthenticated attackers can recursively enumerate arbitrary server filesystem directories and use HTTP...
PT-2026-63347
Name of the Vulnerable Software and Affected Versions n8n versions prior to 1.123.64 Description Custom HTTP header credentials used in LLM sub-nodes, such as OpenAI, Anthropic, and Lemonade, are not properly masked during execution. While these values appear masked in the user interface, they ar...
Taming the Security-Energy Paradox: A Green AI Approach to Optimized Android Malware Detection
An increase in advanced Android malware requires the use of deep learning models, which can run on Android devices. But there is a trade-off between security and energy use, as strong detection models can drain the battery of devices fast. This work tests different Multi-Layer Perceptron MLP mode...
Defense against LLM Backdoors Using Critical Neuron Isolation Pruning
Large language models LLMs are vulnerable to backdoor attacks, where hidden triggers induce malicious outputs. Existing defenses generally fall into inference-time detection or training-time mitigation, but face two key limitations. First, they focus on fine-tuning-based backdoors e.g., PEFT...
Oracle Linux 10 : libvirt (ELSA-2026-18326)
The remote Oracle Linux 10 host has packages installed that are affected by a vulnerability as referenced in the ELSA-2026-18326 advisory. 11.10.0-12.4.0.1 - Set SOURCEDATEEPOCH from changelog Orabug: 32019554 11.10.0-12.4.el102 - cpuconf: Introduce virCPUDefSortFeatures RHEL-180449 -...
Adobe Experience Manager XSS
The vulnerability of the Adobe Media Encoder application relates to the lack of authentication for critical functions. Exploiting this vulnerability allows a malicious actor to perform cross-site scripting attacks XSS using the DOM...
Adobe Experience Manager XSS
The vulnerability of the Adobe Media Encoder application relates to the lack of authentication for critical functions. Exploiting this vulnerability allows a malicious actor to perform cross-site scripting attacks XSS using the DOM...
CVE-2026-65315
Ollama HEAD f0078ae contains an uncontrolled memory allocation vulnerability in the GGUF metadata parser that allows remote attackers to crash the server by supplying a crafted GGUF file with attacker-controlled length and count fields in string lengths, tensor dimension counts, and metadata arra...
CVE-2026-65315 Ollama Remote Denial of Service via Attacker-Controlled Allocation in GGUF Metadata Parser
Ollama HEAD f0078ae contains an uncontrolled memory allocation vulnerability in the GGUF metadata parser that allows remote attackers to crash the server by supplying a crafted GGUF file with attacker-controlled length and count fields in string lengths, tensor dimension counts, and metadata arra...
CVE-2026-65315
Ollama HEAD f0078ae contains an uncontrolled memory allocation vulnerability in the GGUF metadata parser that allows remote attackers to crash the server by supplying a crafted GGUF file with attacker-controlled length and count fields in string lengths, tensor dimension counts, and metadata arra...
CVE-2026-63764 LMDeploy Server-Side Request Forgery via HTTP Redirect Bypass
LMDeploy through 0.14.0, fixed in commit 03c3130, contains a server-side request forgery SSRF vulnerability in the loadhttpurl function within the connection.py media handler, where the private-IP guard validates only the original URL without re-validating hosts after HTTP redirects. An...
Awesome-MoAI-Security — Updated!
Awesome Mobile On-Device AI Security SoK: Attack and Defense Landscape of Mobile On-device AI Systems Yujin Huang, Xin Zheng, Xingliang Yuan, Kwok-Yan Lam Mobile on-device AI systems execute AI models locally through ML frameworks such as LiteRT/TFLite , Core ML , ExecuTorch , ONNX , and...
apidriver (>=0.6.5, <=0.9.1), classification-model (0.0.1) +27 more potentially affected by CVE-2026-59890 via setuptools (56.0.0)
setuptools PYPI version 56.0.0 is affected by a known vulnerability. The following packages have a transitive dependency on setuptools and may be impacted: - apidriver =0.6.5, =3.33.0, =0.41.1, =10.0.0, =0.0.3.6, =1.2.3, =1.0.0, =0.0.18, =1.0.0, =1.4.13, =1.4.29 and more Source cves: CVE-2026-598...
CVE-2026-47395 PraisonAI CLI automatically resolves @url mentions in prompt text and can read loopback URLs into model context
PraisonAI is a multi-agent teams system. Prior to version 4.6.40 of PraisonAI, corresponding to version 1.6.40 of praisonaiagents, PraisonAI's direct-prompt CLI automatically expands @url: mentions in raw prompt text before agent execution begins. If a prompt contains @url:, the CLI calls...
CVE-2026-47395 PraisonAI CLI automatically resolves @url mentions in prompt text and can read loopback URLs into model context
PraisonAI is a multi-agent teams system. Prior to version 4.6.40 of PraisonAI, corresponding to version 1.6.40 of praisonaiagents, PraisonAI's direct-prompt CLI automatically expands @url: mentions in raw prompt text before agent execution begins. If a prompt contains @url:, the CLI calls...
CVE-2026-47395
CVE-2026-47395 – PraisonAI CLI loopback/SSRF via @url: mentions Prior to PraisonAI v4.6.40 (and praisonaiagents v1.6.40), the direct-prompt CLI automatically expands @url:(http|https) prompts, performing an unrestricted urllib.request.urlopen() to the attacker-controlled URL. The fetched response...
CVE-2026-47395 PraisonAI CLI automatically resolves @url mentions in prompt text and can read loopback URLs into model context
PraisonAI is a multi-agent teams system. Prior to version 4.6.40 of PraisonAI, corresponding to version 1.6.40 of praisonaiagents, PraisonAI's direct-prompt CLI automatically expands @url: mentions in raw prompt text before agent execution begins. If a prompt contains @url:, the CLI calls...
