Nvidia counters insider attacks with digital fingerprinting

Nvidia announced today that a digital lab playground for its latest security products is now available, allowing users to try out an artificial intelligence-powered system designed to monitor individual user accounts for potentially dangerous behavior.

The idea, according to the company, is to take the vast amounts of data that many organizations compile on their systems anyway about login and data access events, and use that data to train an AI to monitor user accounts for deviations from their usual patterns. The system moves security teams from a scenario where they have to comb through potentially millions of events each week to identify issues, to the handful of “high risk” events that the system identifies.

Digital Fingerprint ‘Lego Kit’

In an official blog post, Nvidia said the solution is essentially “a sort of Lego kit” for digital fingerprinting, allowing users to customize a solution that best suits their needs.

The system has the potential to reduce the workload and identify potentially malicious elements more quickly, Justin Boitano, Nvidia’s vice president of enterprise and edge computing, said in a briefing.

“This digital fingerprint API workflow enables security organizations to detect threats immediately,” he said. “This means that identity attacks, the most common form of enterprise security attack, can be quickly mitigated and detected immediately by bad actors through social engineering or phishing attempts to obtain employee credentials.”

Nvidia says such systems are just the beginning of AI-powered security efforts. In the future, more detailed user profiles, including data as granular as typing speed and accuracy, could be used to help identify suspicious behavior, the company said in its blog post.

“Data about cyber incidents is gold for building cyber-hardened AI models, but no one wants to share details about real users and intrusions,” Nvidia wrote. “Synthetic data generated from digital fingerprinting variants can fill in the gaps , allowing users to create data that fits their use case.”

The digital fingerprinting system is now available on Nvidia’s AI Enterprise platform, while the new lab module — which allows users to complete real-world deployments on an Equinix-hosted example architecture and interact with Nvidia experts — is available starting today on the company’s LaunchPad website gets.

Copyright © 2023 IDG Communications, Inc.

Source link