AI Sec Watch

InvokeAI versions 5.3.1 through 5.4.2 contain a remote code execution vulnerability (the ability for attackers to run commands on a system they don't own) in the model installation API. The flaw comes from unsafe deserialization (converting data back into usable code without checking if it's trustworthy) of model files using torch.load, which allows attackers to hide malicious code in model files that gets executed when loaded.

In gpt_academic version 3.83 and earlier, the CodeInterpreter plugin has a vulnerability where prompt injection (tricking an AI by hiding instructions in its input) allows attackers to inject malicious code. Because the application executes LLM-generated code without a sandbox (a restricted environment that isolates code from the main system), attackers can achieve RCE (remote code execution, where an attacker can run commands on a system they don't own) and potentially take over the backend server.

Applio, a voice conversion tool, has a vulnerability in versions 3.2.8-bugfix and earlier where it unsafely deserializes (converts untrusted data back into code objects) user-supplied model file paths using torch.load, which can allow attackers to run arbitrary code on the system. The vulnerability exists in the inference.py and tts.py files, where user input is passed directly to functions that load models without proper validation.

Applio, a voice conversion tool, has a vulnerability in versions 3.2.8-bugfix and earlier where it unsafely deserializes (reconstructs objects from stored data without validation) user-supplied model files using `torch.load`, which could allow attackers to run arbitrary code on the affected system.

Applio, a voice conversion tool, has a vulnerability in versions 3.2.8-bugfix and earlier where it unsafely deserializes (converts untrusted data back into objects) user-supplied model files using `torch.load`, potentially allowing attackers to run arbitrary code on affected systems.

CVE-2025-29783 is a remote code execution vulnerability in vLLM (a software engine for running large language models efficiently) that occurs when it is configured with Mooncake, a distributed system component. Attackers can exploit unsafe deserialization (the process of converting stored data back into usable objects) exposed over ZMQ/TCP (network communication protocols) to run arbitrary code on any connected systems in a distributed setup.

vLLM, a system for running large language models efficiently, uses the outlines library to support structured output (guidance on what format the AI's answer should follow). The outlines library stores compiled grammar rules in a cache on the hard drive, which is turned on by default. A malicious user can send many requests with different output formats, filling up this cache and causing the system to run out of disk space, making it unavailable to others (a denial of service attack). This problem affects only the V0 engine version of vLLM.

SmartOS, a hypervisor (virtualization software that manages virtual machines) used in Triton Data Center and other products, contains static host SSH keys (unchanging cryptographic credentials for remote access) in a specific Debian 12 LX zone image from July 2024. This means multiple systems could potentially share the same SSH keys, allowing unauthorized remote access if someone obtains these keys.

Researchers have discovered advanced data smuggling techniques using invisible Unicode characters (invisible text that computers can process but humans cannot see) to hide information in LLM inputs and outputs. The technique, called Sneaky Bits, can encode any character or sequence of bytes using only two invisible characters, building on earlier methods that used Unicode Tags and Variant Selectors to inject hidden instructions into AI systems.