Inside Advanced Scale Challenges|Saturday, October 1, 2016
  • Subscribe to EnterpriseTech Weekly Updates: Subscribe by email

Cloud Security Gets a Boost From Deep Learning 

Security-dark

A range of new cyber security approaches is emerging that seek to shield cloud applications from attack. Some emphasize a human intelligence approach as a way of staying ahead of threats. Others are embracing artificial intelligence techniques in order to deliver real-time detection and prevention of cyber threats.

The latest AI approach comes from partners Deep Instinct and FireLayers, who on Wednesday (Feb. 17) unveiled a cyber defense platform for enterprise cloud applications based on deep learning technology. Deep Instinct, Tel Aviv, Israel, provides AI technology that "learns" to detect cyber threats so it can better predict future attacks. The system runs on a policy-based platform from FireLayers designed to monitor cloud applications and data.

The combination of deep learning technology and FireLayer's security gateway targets customers moving applications to the cloud. If the predictive deep learning technology detects a malicious file, the security platform enforces compliance with security policies to prevent the distribution of malware or infection.

Others have taken a more hands-on approach to cyber security, focusing instead on a virtualized "intelligence-led" threat management platform designed to provide companies with real-time threat updates. In January, IT security specialist FireEye Inc. (NASDAQ: FEYE) announced a deal to acquire privately held iSight Partners. Dallas-based iSight Partners reportedly hires overseas agents to monitor online forums to stay ahead of potential cyber threats. Its human intelligence approach differs from cyber defense firms like Deep Instinct that rely heavily on data and algorithms to fend off attacks.

The human intelligence-led capability would be integrated with FireEye's "virtual execution engine" that used threat intelligence to identify and block cyber attacks as they unfold.

These and other approaches appear to be convincing more enterprise that they can securely move data and applications to the cloud. According a cloud adoption study released last week, security is increasingly seen as a "primary benefit" to shifting from on-premises to cloud platforms.

"Organizations no longer have to choose between the cloud’s convenience and cyber security," Deep Instinct CEO Guy Caspi asserted in a statement announcing the partnership with FireLayers. The Israeli company touts its deep learning predictive capabilities as protecting operating systems, servers and mobile devices against advanced persistent attacks and other cyber threats.

The use of deep learning technology is a new twist in the effort to provide enterprise cloud adopters with real-time threat data that can be used to block attacks as they unfold. The partners said deep learning is deployed via the FireLayers security platform to spot "first-seen" attacks. Deep Instinct claims its proprietary deep learning algorithms can train the cyber defense system to predict future and unknown cyber attacks.

The approach works by breaking down data objects into their constituent parts. The process is said to be agnostic in terms of data origin. Based on this metadata, the deep learning algorithms train the system to make predictions about future threats.

The partnership with FireLayers boosts Deep Instinct's ability to distribute the cyber threat technology across enterprise IT infrastructure.

 

About the author: George Leopold

George Leopold has written about science and technology for more than 25 years, focusing on electronics and aerospace technology. He previously served as Executive Editor for Electronic Engineering Times.

Add a Comment

Share This