What changed in AI security, Mar 17 to Mar 23, 2025
Mar 17 to Mar 23, 2025 (ISO week 2025-W12). Weeks run Monday to Sunday in UTC.
51 records published, +45 on the previous week: 51 vulnerabilities (+46), 0 incidents (no change), 0 research items (no change), 0 news items (-1), 0 policy items (no change).
Critical and high advisories
Vulnerability records rated critical or high, newest first. Showing 25 of 37.- High
CVE-2025-1473: mlflow Cross-Site Request Forgery in the Signup feature
CVE-2025-1473NVD/CVE Database - High
CVE-2025-0453: mlflow denial of service through the /graphql endpoint
CVE-2025-0453NVD/CVE Database - High
CVE-2025-0317: Ollama denial of service via crafted GGUF model file upload
CVE-2025-0317NVD/CVE Database - High
CVE-2025-0315: Ollama denial of service through crafted GGUF model file creation
CVE-2025-0315NVD/CVE Database - High
CVE-2025-0312: Ollama denial of service through crafted GGUF model file upload
CVE-2025-0312NVD/CVE Database - Critical
CVE-2024-9070: BentoML runner server deserialization flaw enabling arbitrary code execution
CVE-2024-9070NVD/CVE Database - Critical
CVE-2024-9053: vllm-project vllm RPC server remote code execution via unsafe deserialization
CVE-2024-9053NVD/CVE Database - High
CVE-2024-8966: Gradio denial of service through multipart boundary processing in file upload
CVE-2024-8966NVD/CVE Database - High
CVE-2024-8859: mlflow/mlflow path traversal in dbfs service via URL file protocol
CVE-2024-8859NVD/CVE Database - High
CVE-2024-8063: Ollama divide by zero via crafted block_count in GGUF model import
CVE-2024-8063NVD/CVE Database - High
CVE-2024-7959: open-webui SSRF through the /openai/models endpoint
CVE-2024-7959NVD/CVE Database - Critical
CVE-2024-7776: onnx download_model arbitrary file overwrite via path traversal in tar files
CVE-2024-7776NVD/CVE Database - High
CVE-2024-12720: huggingface/transformers ReDoS in post_process_single()
CVE-2024-12720NVD/CVE Database - High
CVE-2024-12217: gradio-app/gradio path traversal on Windows via NTFS alternate data streams
CVE-2024-12217NVD/CVE Database - High
CVE-2024-12065: haotian-liu/llava local file inclusion in gradio web UI
CVE-2024-12065NVD/CVE Database - High
CVE-2024-12055: Ollama out-of-bounds read via malicious gguf model file causes denial of service
CVE-2024-12055NVD/CVE Database - Critical
CVE-2024-11041: vllm-project vllm remote code execution in MessageQueue.dequeue() via pickle
CVE-2024-11041NVD/CVE Database - High
CVE-2024-11037: binary-husky/gpt_academic path traversal reads config.py on Windows
CVE-2024-11037NVD/CVE Database - High
CVE-2024-11031: gpt_academic server-side request forgery in Markdown_Translate API
CVE-2024-11031NVD/CVE Database - High
CVE-2024-11030: GPT Academic server-side request forgery through HotReload plugin
CVE-2024-11030NVD/CVE Database - High
CVE-2024-10940: langchain-core arbitrary file read through ImagePromptTemplate input variables
CVE-2024-10940NVD/CVE Database - High
CVE-2024-10707: gaizhenbiao/chuanhuchatgpt local file inclusion through handle_dataset_selection
CVE-2024-10707NVD/CVE Database - High
CVE-2024-10650: ChuanhuChatGPT unauthenticated denial of service via multipart boundary payloads
CVE-2024-10650NVD/CVE Database - High
GHSA-879v-fggm-vxw2: LiteLLM Has a Leakage of Langfuse API Keys
CVE-2025-0330GitHub Advisory Database - High
GHSA-fjcf-3j3r-78rp: LiteLLM Has an Improper Authorization Vulnerability
CVE-2025-0628GitHub Advisory Database
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-2025-0317: Ollama denial of service via crafted GGUF model file upload CVE-2025-0317NVD/CVE Database | Not listed | 14.2% | |
| CVE-2025-0453: mlflow denial of service through the /graphql endpoint CVE-2025-0453NVD/CVE Database | Not listed | 11.0% |
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.| Package | Ecosystem | LLM SDKs | Release | Released |
|---|---|---|---|---|
| @mem0/community | npm | LangChain | 0.0.1 | |
| flash-linear-attention | PyPI | Hugging Face Hub / Transformers | 0.1 | |
| google-adk | PyPI | Google Agent Development Kit | 0.0.1 |
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 | 0.3 | +9.8 |
Research
Peer-reviewed first, then newest.No research papers or reports were published in this week.
Policy and regulation
Newest first.No regulatory or policy records were published in this week.
Generated from the AI Sec Watch database at . Every item links to its record.