AI Sec Watch

MLflow version 2.11.1 has a vulnerability where attackers can create multiple models with the same name by using URL encoding (a technique that converts special characters into a format safe for web addresses). This allows attackers to cause denial of service (making a service unavailable) or data poisoning (inserting corrupted or malicious data), where an authenticated user might accidentally use a fake model instead of the real one because the system treats URL-encoded and regular names as different.

A Server-Side Request Forgery vulnerability (SSRF, a flaw that lets attackers trick a server into making requests to unintended targets) exists in langchain version 0.1.5's Web Research Retriever component, which fails to block requests to local network addresses. This allows attackers to scan ports, access local services, read cloud metadata, and potentially execute arbitrary code (run commands on a system they don't own) by exploiting internal APIs.

A Local File Inclusion vulnerability (LFI, a flaw that lets attackers read files they shouldn't access) was found in MLflow version 2.9.2. The bug exists because the application doesn't properly check the fragment part of web addresses (the section after the '#' symbol) for directory traversal sequences like '../', which allow attackers to navigate folders and read sensitive files like system password files.

MLflow version 8.2.1 has a command injection vulnerability (a flaw where attackers can execute arbitrary commands by inserting malicious code into a system command) in its HTTP dataset loading function. When loading datasets, the software doesn't properly clean up filenames from URLs, allowing attackers to write files anywhere on the system and potentially run harmful commands.

PyTorch Lightning version 2.2.1 has a critical vulnerability where attackers can execute arbitrary code on self-hosted applications by crafting malicious serialized data (deepdiff.Delta objects, which are used to represent changes to data). The vulnerability exists because the application doesn't properly block access to dunder attributes (special Python attributes starting with underscores), allowing attackers to bypass security restrictions and modify the application's state.

Gradio version 4.25 has a local file inclusion vulnerability (a security flaw where attackers can read files they shouldn't access) in its JSON component. The problem occurs because the `postprocess()` function doesn't properly validate user input before parsing it as JSON, and if the JSON contains a `path` key, the system automatically moves that file to a temporary directory where attackers can retrieve it using the `/file=..` endpoint.

A Server-Side Request Forgery vulnerability (SSRF, where a server can be tricked into making requests to unintended locations) exists in Gradio version 4.21.0 in the `/queue/join` endpoint and `save_url_to_cache` function. The vulnerability occurs because user-supplied URL input is not properly validated before being used to make HTTP requests, allowing attackers to access internal networks or sensitive cloud server information.

EmailGPT has a prompt injection vulnerability (a technique where attackers hide malicious instructions in their input to trick the AI) that allows anyone with access to the service to manipulate it into leaking its internal system prompts or executing unintended commands. Attackers can exploit this by submitting specially crafted requests that trick the service into providing harmful information or performing actions it wasn't designed to do.

A workflow file (a set of automated tasks) in the Gradio project has a security flaw where it runs code from external copies of the repository without proper safety checks, allowing attackers to steal sensitive secrets (like API keys and authentication tokens). This happens because the workflow trusts and executes code from forks (unauthorized copies of the project) in an environment that has access to the main repository's secrets.