Cybersecurity researchers have disclosed a critical security flaw in the Lightning AI Studio development platform that, if successfully exploited, could allow for remote code execution.
The vulnerability, rated a CVSS score of 9.4, enables “attackers to potentially execute arbitrary commands with root privileges” by exploiting a hidden URL parameter, application security firm Noma said in a report shared with The Hacker News.
“This level of access could hypothetically be leveraged for a range of malicious activities, including the extraction of sensitive keys from targeted accounts,” researchers Sasi Levi, Alon Tron, and Gal Moyal said.
Protect your privacy by Mullvad VPN. Mullvad VPN is one of the famous brands in the security and privacy world. With Mullvad VPN you will not even be asked for your email address. No log policy, no data from you will be saved. Get your license key now from the official distributor of Mullvad with discount: SerialCart® (Limited Offer).
➤ Get Mullvad VPN with 12% Discount
The issue is embedded in a piece of JavaScript code that could facilitate unfettered access to a victim’s development environment, as well as run arbitrary commands on an authenticated target in a privileged context.
Noma said it found a hidden parameter called “command” in user-specific URLs – e.g., “lightning.ai/PROFILE_USERNAME/vision-model/studios/STUDIO_PATH/terminal?fullScreen=true&commmand=cmVzc…” – which could be used to pass a Base64-encoded instruction to be executed on the underlying host.
Even worse, the loophole could be weaponized to run commands that can exfiltrate critical information such as access tokens and user information to an attacker-controlled server.
Successful exploitation of the vulnerability means that it could permit an adversary to execute arbitrary privileged commands and gain root access, harvest sensitive data, and manipulate the file system to create, delete, or modify files on the server.
All an attacker needs to pull this off is prior knowledge of a profile username and their associated Lightning AI Studio, details that are publicly available via the Studio templates gallery.
Armed with this information, the threat actor can then craft a malicious link such that it triggers code execution on the identified Studio under root permissions. Following responsible disclosure on October 14, 2024, the problem has been resolved by the Lightning AI team as of October 25.
“Vulnerabilities like these underscore the importance of mapping and securing the tools and systems used for building, training, and deploying AI models because of their sensitive nature,” the researchers said.
Found this article interesting? Follow us on Twitter and LinkedIn to read more exclusive content we post.
Some parts of this article are sourced from:
thehackernews.com
SOC Analysts – Reimagining Their Role Using AI