Security

Patches

Nvidia's newborn ChatRTX bot patched for security bugs

Flaws enable privilege escalation and remote code execution


Nvidia's AI-powered ChatRTX app launched just six week ago but already has received patches for two security vulnerabilities that enabled attack vectors, including privilege escalation and remote code execution.

ChatRTX, formerly known as Chat with RTX, was launched in February to provide Nvidia GPU owners with an AI chatbot that could run locally on RTX 30 and 40-series hardware with at least 8 GB of VRAM. While this solution couldn't promise as much power as a cloud-based alternative, being able to run it locally has been an upside for early users.

One of the downsides for users of earlier versions was that it harbored two security bugs designated CVE‑2024‑0082 and CVE‑2024‑0083. These flaws existed in all versions of ChatRTX up to version 0.2. The latter is rated at a medium severity level of 6.5, while the former is an 8.2 high-level problem.

CVE‑2024‑0083 could allow attackers to perform denial of service attacks, steal data, and even perform remote code execution (RCE). A score of 6.5 for these issues is relatively tame, and many others can score more than 9 points or even the maximum 10 out of 10 in the case of the Atlassian Confluence RCE exploit.

The other vulnerability, CVE‑2024‑0082, enables data stealing (again), data tampering, and even privilege escalation. This issue may have warranted the higher severity score since privilege escalation can render a computer totally open to intrusion.

RCE combined with privilege escalation could prove potent combo as well. Nvidia says it's possible via open file requests and by causing cross-site scripting errors that then allows browser scripts to be run. It's unknown if anyone was actually compromised thanks to these ChatRTX bugs. We have reached out to Nvidia for comment and will update when we hear back.

All users have to do is update to ChatRTX version 0.2. Confusingly, Nvidia warns that "the version numbers of the last affected version and the updated version are both 0.2" so maybe just completely reinstall ChatRTX to be safe. ®

Send us news
1 Comment

India, Nvidia, discuss jointly developed AI chip

Current capabilities mean local manufacturing is not likely – but a chip tuned to Indian needs could work

Nvidia CEO whines Europeans aren’t buying enough GPUs

EU isn’t keeping up with US and China investments, AI arms dealer says

No-Nvidias networking club convenes in search of open GPU interconnect

Ultra Accelerator Link consortium promises 200 gigabits per second per lane spec will debut in Q1 2025

xAI picked Ethernet over InfiniBand for its H100 Colossus training cluster

Work already underway to expand system to 200,000 Nvidia Hopper chips

AI firms and civil society groups plead for passage of federal AI law ASAP

Congress urged to act before year's end to support US competitiveness

Sorry, but the ROI on enterprise AI is abysmal

Appen points to, among other problems, a lack of high-quality training data labeled by humans

Open source LLM tool primed to sniff out Python zero-days

The static analyzer uses Claude AI to identify vulns and suggest exploit code

The troublesome economics of CPU-only AI

At the end of the day, it all boils down to tokens per dollar

Hugging Face puts the squeeze on Nvidia's software ambitions

AI model repo promises lower costs, broader compatibility for NIMs competitor

Voice-enabled AI agents can automate everything, even your phone scams

All for the low, low price of a mere dollar

Anthropic's latest Claude model can interact with computers – what could go wrong?

For starters, it could launch a prompt injection attack on itself...

Cast a hex on ChatGPT to trick the AI into writing exploit code

'It was like watching a robot going rogue' says researcher