AI
Breaking Wire
Content
Ecosystem
Community
Pricing
AI Vulnerability Database
916 vulnerabilities tracked — CVEs, prompt injection, model exploits.
HIGH
CVE-2024-0090
NVIDIA GPU Driver Use-After-Free Flaw Allows Privilege Escalation in Multi-Tenant ML Environments
NVIDIA GPU Display Driver for Linux (versions < 555.48.07)
NVIDIA Data Center GPU Manager (DCGM)
18 Nov 2025
6 views
Patched
HIGH
Previous
Page 40 of 77
Next
CVE-2025-45882
Cross-Tenant Data Leakage in AWS Bedrock's Custom Model Fine-Tuning Service
AWS Bedrock
18 Nov 2025
10 views
Patched
CRITICAL
Cross-Tenant Data Access in Azure AI Services via 'a la carte' API Manipulation
Azure Machine Learning
Azure Cognitive Search
Azure OpenAI Service
18 Nov 2025
24 views
Patched
MEDIUM
CVE-2025-48175
Denial of Service in NVIDIA Triton Inference Server via Malformed Model Input Tensor
NVIDIA Triton Inference Server < 2.58.0
15 Nov 2025
0 views
Patched
CRITICAL
Malicious PyPI Packages Typosquatting ML Libraries Steal Developer Credentials
Python developers using PyPI
15 Nov 2025
0 views
Unpatched
HIGH
GitHub Copilot Workspace Agent Vulnerable to Command Injection via Malicious Markdown
GitHub Copilot Workspace (Beta versions < 0.8.0)
15 Nov 2025
0 views
Patched
HIGH
Stealthy Indirect Prompt Injection in Email Assistants via Unicode Obfuscation
Various AI Email Assistants
LLM Agent Frameworks
15 Nov 2025
0 views
Unpatched
HIGH
Indirect Prompt Injection in Email-Based Support Bot Leads to PII Exfiltration
Custom AI agents processing external data
AI-powered helpdesk assistants
15 Nov 2025
0 views
Unpatched
HIGH
CVE-2025-48211
NVIDIA GPU Display Driver Kernel Mode Layer Vulnerability Leading to Privilege Escalation
NVIDIA GPU Display Driver (Windows) < 545.31
NVIDIA GPU Display Driver (Linux) < 545.87
14 Nov 2025
0 views
Patched
HIGH
CVE-2025-45219
NVIDIA Triton Inference Server Model Repository Path Traversal
NVIDIA Triton Inference Server < 2.45.0
12 Nov 2025
0 views
Patched
MEDIUM
AIM Character-Roleplaying Jailbreak Bypasses Safety Filters on Major LLMs
OpenAI GPT-4 Series
Anthropic Claude 3 Series
Google Gemini Pro
12 Nov 2025
0 views
Unpatched
MEDIUM
Data Exfiltration via Adversarially Crafted Images in AWS Bedrock's Claude 3 Sonnet API
AWS Bedrock (Claude 3 Sonnet model)
Anthropic Claude 3 API
GCP Vertex AI (Gemini Pro Vision model)
12 Nov 2025
5 views
Unpatched