aisecwatch.com
DashboardVulnerabilitiesNewsResearchArchiveStatsDatasetFor devs
Subscribe
aisecwatch.com

Real-time AI security monitoring. Tracking AI-related vulnerabilities, safety and security incidents, privacy risks, research developments, and policy changes.

Navigation

VulnerabilitiesNewsResearchDigest ArchiveNewsletter ArchiveSubscribeData SourcesStatisticsDatasetAPIIntegrationsWidgetRSS Feed

Maintained by

Truong (Jack) Luu

Information Systems Researcher

AI & LLM Vulnerabilities

Security vulnerabilities, privacy incidents, safety concerns, and policy updates affecting LLMs and AI agents.

to
Export CSV
186 items

GHSA-9cr9-25q5-8prj: PraisonAI vulnerable to unauthenticated arbitrary file read via MCP workflow.show, workflow.validate, deploy.validate

highvulnerability
security
May 29, 2026
CVE-2026-47394

PraisonAI has a security vulnerability where unauthenticated users can read any file on the system through multiple handlers like `workflow.show`, `workflow.validate`, and `deploy.validate`. The problem exists because these file-handling tools don't validate or restrict file paths before reading them, and the dispatcher that calls these tools doesn't enforce security checks on incoming requests.

GitHub Advisory Database

GHSA-78r8-wwqv-r299: PraisonAI: Arbitrary code execution via unguarded `spec.loader.exec_module` in `agents_generator.py` - sibling of CVE-2026-44334

highvulnerability
security
May 29, 2026
CVE-2026-47398

PraisonAI versions up to 4.6.37 contain a critical vulnerability where two functions in `agents_generator.py` use `spec.loader.exec_module` (a method that loads and runs Python code from a file) without any security checks or validation. An attacker can exploit this by providing a malicious Python file path through YAML configuration, either by modifying shared config files, submitting a crafted recipe via GitHub, or using prompt injection (tricking the AI into loading code from a malicious path), resulting in arbitrary code execution (the ability to run any commands on the system).

CVE-2026-4944: vllm-project/vllm version 0.14.1 contains a vulnerability where the `trust_remote_code=True` parameter is hardcoded in t

highvulnerability
security
May 28, 2026
CVE-2026-4944

vLLM version 0.14.1 has a security flaw where a setting called `trust_remote_code` is permanently turned on in two model files, even when users try to turn it off. This allows RCE (remote code execution, where attackers can run harmful code on your computer) through malicious models downloaded from HuggingFace, a popular model repository. This is a partial fix attempt for two earlier vulnerabilities that didn't fully solve the problem.

CVE-2026-48545: Gradio before version 6.15.0 contains a cookie injection vulnerability that allows remote attackers to perform cross-Spa

mediumvulnerability
security
May 27, 2026
CVE-2026-48545

Gradio versions before 6.15.0 have a cookie injection vulnerability that lets attackers perform session fixation (tricking a system into using a fake session ID) across multiple user spaces. An attacker controlling one Gradio Space can inject a cookie into a shared HTTP client (a tool that sends web requests) that automatically gets sent to all other legitimate Spaces, affecting every user on that Gradio deployment.

CVE-2026-44895: GitLab MCP Server lets an AI agent talk directly to GitLab. Prior to 0.6.0, the HTTP transport in src/transport.ts ships

criticalvulnerability
security
May 26, 2026
CVE-2026-44895

GitLab MCP Server (a tool that lets AI agents interact with GitLab) had a critical security flaw in versions before 0.6.0 where its HTTP transport exposed an unauthenticated endpoint (a service that processes requests without checking who is calling it) to any website, combined with a misconfiguration that made it accessible from all network interfaces instead of just locally. This allowed attackers from anywhere to make changes to GitLab repositories using the server operator's stored credentials.

GHSA-m549-qq94-fvhg: LMDeploy: Arbitrary code execution via hardcoded trust_remote_code=True in lmdeploy model initialization

highvulnerability
security
May 21, 2026
CVE-2026-46432

LMDeploy, a model serving tool, hardcodes `trust_remote_code=True` (a setting that allows executing custom Python code from downloaded models) when loading models from HuggingFace. An attacker who can control which model path the system loads could point it to a malicious model repository, causing arbitrary code execution (running any commands they want) with the privileges of the LMDeploy server process. This affects LMDeploy version 0.12.3 and earlier.

GHSA-7wx4-6vff-v64p: Diffusers: TOCTOU Trust Remote Code Bypass

highvulnerability
security
May 20, 2026
CVE-2026-45804

The `diffusers` package has a TOCTOU (time-of-check-time-of-use, where a security check happens at one moment but the actual data used comes from a different moment) vulnerability in its `DiffusionPipeline.from_pretrained` function that loads models from HuggingFace Hub. An attacker can bypass the `trust_remote_code` security check by updating a repository between two separate download calls, allowing arbitrary code to execute without the user explicitly approving it.

CVE-2026-4137: In mlflow/mlflow versions prior to 3.11.0, the `get_or_create_nfs_tmp_dir()` function in `mlflow/utils/file_utils.py` cr

criticalvulnerability
security
May 18, 2026
CVE-2026-4137

MLflow versions before 3.11.0 create temporary directories with overly permissive access permissions (world-writable or group-writable), allowing local attackers to modify model files and execute arbitrary code when those files are loaded. This is especially dangerous in shared environments like Databricks where multiple users access the same network storage.

CVE-2026-45829: A pre-authentication, code injection vulnerability in version 1.0.0 or later of the ChromaDB Python project allows an un

criticalvulnerability
security
May 18, 2026
CVE-2026-45829

ChromaDB (a Python project for storing AI embeddings) versions 1.0.0 and later contain a code injection vulnerability that lets unauthenticated attackers run arbitrary code on the server by sending a malicious model repository with a specific setting enabled to a particular API endpoint. The vulnerability has a CVSS score (a 0-10 severity rating) of 10.0, marking it as critical.

CVE-2026-42463: SQLBot is an intelligent Text-to-SQL system based on large language models and RAG. Prior to 1.8.0, SQLBot contains a Cr

highvulnerability
security
May 13, 2026
CVE-2026-42463

SQLBot is a Text-to-SQL system (software that converts natural language questions into database queries) that uses large language models and RAG (retrieval-augmented generation, where the AI pulls in external documents to answer questions). Before version 1.8.0, it had an IDOR vulnerability (insecure direct object reference, where an attacker can access resources belonging to other users by manipulating request parameters), allowing attackers to access and modify database schemas and data from other workspaces or organizations.

CVE-2026-31239: The mamba language model framework thru 2.2.6 is vulnerable to insecure deserialization (CWE-502) when loading pre-train

criticalvulnerability
security
May 12, 2026
CVE-2026-31239

The mamba language model framework (versions up to 2.2.6) has a vulnerability in how it loads pre-trained models from HuggingFace Hub (a platform where AI models are shared). When loading models, it uses an unsafe method called torch.load() without the weights_only=True security parameter, which allows attackers to sneak malicious code into model files. An attacker could upload a compromised model to HuggingFace Hub, and when someone downloads and loads it, the attacker's code runs on their computer.

CVE-2026-31229: The Adversarial Robustness Toolbox (ART) thru 1.20.1 contains an insecure deserialization vulnerability (CWE-502) in its

criticalvulnerability
security
May 12, 2026
CVE-2026-31229

The Adversarial Robustness Toolbox (ART) version 1.20.1 and earlier has a vulnerability in how it loads AI model files, specifically in its Kubeflow component (a system for running machine learning workflows). When loading model weights using torch.load() without the weights_only=True security parameter, the software deserializes arbitrary Python objects via Pickle (a Python serialization library), allowing attackers to execute malicious code by uploading a crafted model file or manipulating the model location parameter.

CVE-2026-31228: The Adversarial Robustness Toolbox (ART) thru 1.20.1 contains a remote code execution vulnerability in its Kubeflow comp

criticalvulnerability
security
May 12, 2026
CVE-2026-31228

The Adversarial Robustness Toolbox (ART) version 1.20.1 and earlier has a remote code execution (RCE, where an attacker can run commands on a system they don't own) vulnerability in its Kubeflow component. The vulnerability exists because the robustness evaluation function uses eval() (a function that executes text as Python code) without checking user input, allowing an attacker to submit malicious Python code that runs on the system when the evaluation function processes it.

CVE-2026-31224: The snorkel library thru v0.10.0 contains an insecure deserialization vulnerability (CWE-502) in the MultitaskClassifier

criticalvulnerability
security
May 12, 2026
CVE-2026-31224

The snorkel library (a tool for machine learning data labeling) versions up to 0.10.0 has a security flaw in its MultitaskClassifier.load() method that allows arbitrary code execution (running any commands an attacker wants on your computer). The problem occurs because the method uses torch.load() without the weights_only=True security setting, which means it can deserialize (reconstruct) malicious Python objects from model files that an attacker provides.

CVE-2026-31222: The snorkel library thru v0.10.0 contains an insecure deserialization vulnerability (CWE-502) in the Trainer.load() meth

highvulnerability
security
May 12, 2026
CVE-2026-31222

The snorkel library up to version 0.10.0 has a vulnerability in its Trainer.load() method that unsafely deserializes (converts saved data back into objects) model files using torch.load() without security protections. An attacker can craft a malicious model file that executes arbitrary code (RCE, remote code execution) when a user loads it with this method.

CVE-2026-42345: FastGPT is an AI Agent building platform. In versions 4.14.11 and prior, FastGPT's isInternalAddress() function in packa

highvulnerability
security
May 8, 2026
CVE-2026-42345

FastGPT, an AI platform for building AI agents, has a vulnerability in versions 4.14.11 and earlier where its isInternalAddress() function fails to block access to cloud metadata endpoints (services that store sensitive system information). Attackers can bypass the blocklist using URL encoding techniques (methods to disguise URLs), and because a security check is disabled by default, the metadata endpoint remains accessible without additional protection.

GHSA-4vg5-rp28-gvjf: Open WebUI has Improper Authorization Control

highvulnerability
security
May 8, 2026
CVE-2026-44567

Open WebUI version 0.1.105 has an authorization flaw where users with a 'pending' status (the default for new sign-ups) can still receive authentication tokens (JWT, a standardized way to prove identity in web requests) and make API calls meant only for approved users, bypassing the authorization check (the system that verifies what actions a user is allowed to perform) that is only enforced on the website interface, not in the backend API itself.

GHSA-98h9-4798-4q5v: Diffusers has a `trust_remote_code` bypass via `custom_pipeline` and local custom components

highvulnerability
security
May 7, 2026
CVE-2026-44513

Diffusers, a popular AI library, had a security flaw where the `trust_remote_code` parameter (a safety check to prevent running untrusted code) could be bypassed in three ways when loading models with `DiffusionPipeline.from_pretrained()`. An attacker could execute arbitrary code on a user's machine even when the user explicitly set `trust_remote_code=False` or left it at its default safe setting. The vulnerability affected users loading custom pipelines (external code) or local model snapshots (saved model files).

GHSA-j7w6-vpvq-j3gm: Duplicate Advisory: Diffusers has a `trust_remote_code` bypass via `custom_pipeline` and local custom components

highvulnerability
security
May 6, 2026

The Diffusers library has a vulnerability where arbitrary code can be silently executed when loading a pipeline from HuggingFace Hub, bypassing the `trust_remote_code` security check. An attacker can craft a repository with custom code in a Python file that gets automatically executed during `DiffusionPipeline.from_pretrained()` without requiring the `trust_remote_code=True` parameter or any visible warning, allowing remote code execution (RCE, where an attacker runs commands on a system they don't own).

GHSA-mgx6-5cf9-rr43: Keras vulnerable to DoS via Malicious .keras Model (HDF5 Shape Bomb Causes Petabyte Allocation in KerasFileEditor)

highvulnerability
security
May 6, 2026
CVE-2026-0897

Keras has a critical vulnerability in its model loader (KerasFileEditor) that allows attackers to cause a Denial of Service (DoS, where a system becomes unusable) by uploading malicious .keras files. An attacker can craft a small .keras file (100-400 KB) that declares an extremely large dataset shape in its HDF5 weight file (a binary format for storing weights in neural networks), but stores only a few bytes of actual data. When Keras loads this file, it attempts to allocate petabytes of RAM based on the declared shape, immediately crashing the system and killing any applications processing the model.

1 / 10Next
GitHub Advisory Database
NVD/CVE Database

Fix: Update Gradio to version 6.15.0 or later. The vulnerability is fixed in the release available at https://github.com/gradio-app/gradio/releases/tag/gradio%406.15.0.

NVD/CVE Database

Fix: Update GitLab MCP Server to version 0.6.0, which fixes the vulnerability.

NVD/CVE Database
Hugging Face Security Advisories
Hugging Face Security Advisories

Fix: Update MLflow to version 3.11.0 or later.

NVD/CVE Database
NVD/CVE Database

Fix: This vulnerability is fixed in version 1.8.0. Users should upgrade SQLBot to 1.8.0 or later.

NVD/CVE Database
NVD/CVE Database
NVD/CVE Database
NVD/CVE Database
NVD/CVE Database
NVD/CVE Database
NVD/CVE Database
GitHub Advisory Database

Fix: Upgrade to diffusers version 0.38.0 or later by running: `pip install --upgrade "diffusers>=0.38.0"`. The fix moves the `trust_remote_code` security check to `get_cached_module_file()` in `src/diffusers/utils/dynamic_modules_utils.py`, which is the actual point where all dynamic modules are loaded. If immediate upgrading is not possible, the source recommends only using `from_pretrained()` with trusted sources, avoiding `custom_pipeline=` parameters pointing to different repositories without inspecting their code first, and manually checking local snapshots for unexpected `.py` files before loading them, though these are only temporary mitigations and not complete fixes.

Hugging Face Security Advisories
Hugging Face Security Advisories
GitHub Advisory Database