Nvidia releases update for ‘critical’ vulnerabilities in AI stack

Triton, Nvidia’s open-source inference server for optimizing AI model deployment, is now at the center of newly revealed security vulnerabilities.

Technology company Nvidia released a software update on Saturday to patch vulnerabilities in its Triton server, which is used by clients to run artificial intelligence models.

The vulnerabilities, which cybersecurity company Wiz calls “critical,” could lead to the takeover of AI models, data theft and response manipulation if not patched.

“Wiz Research found a chain of vulnerabilities that, when combined, could let an attacker with no prior access take full control of an AI server,” Wiz head of vulnerability research Nir Ohfeld told Cointelegraph.

Read more