Box Announces New Malware Deep Scan Capability in Box Shield To Combat Ransomware

Showcases advanced set of updates for threat detection and auto-classification powered by machine learning

Box, Inc. , the leading Content Cloud, announced new capabilities for Box Shield, the company’s flagship security control and intelligent threat detection solution, to help customers reduce the risk of ransomware by scanning files in near real-time as they are uploaded to Box. The new capabilities, which will be demonstrated later today at BoxWorks 2021, leverage deep learning technology and external threat intelligence to analyze files and stop sophisticated malware before it causes business disruption.

Marketing Technology News: Sensibill Named a Top Growing Company by The Globe and Mail for the Second Consecutive Year

“The number of ransomware attacks surged by 288 percent in the first half of 2021 and this will only increase as more businesses go digital,” said Diego Dugatkin, Chief Product Officer at Box. “Our approach to security is to provide customers with one secure platform to manage and secure their content, and Box Shield brings together user-friendly security controls and intelligent threat detection natively into the Box Content Cloud. By leveraging the latest deep learning technology, we are adding an extra layer of threat detection to Box Shield, making it even easier for IT and security teams to identify malware in near real-time without slowing down work.”

“If it seems like the threat to our data has gotten dramatically worse as of late, it is because it has,” said Frank Dickson, Program Vice President, Security & Trust, IDC. “According to IDC survey data, almost half of U.S. boards of directors have specifically demanded presentations by their CISOs as to how they are approaching ransomware, driving organizations to fortify their security strategy. Box is delivering tools to help, illuminating questionable and malicious content with embedded deep learning scanning technology. Box Shield helps protect users and organizations in near real-time to address the spread of malware without introducing multi-solution, multi-vendor complexity.”

Strengthening malware detection in Box Shield with deep learning

Bolt-on malware detection solutions create friction and disrupt work by quarantining potentially malicious content and triggering cumbersome reviews that slow down the business, ultimately leading to teams working outside of secure processes to get work done. Box Shield eliminates these obstacles by natively embedding malware detection into the Box Content Cloud to deliver both a seamless end user experience and near real-time alerts for IT security teams.

At BoxWorks 2021, Box is extending Box Shield’s detection capabilities to identify more sophisticated malware by adding deep learning technology that complements traditional hash-based or file-fingerprint scanning approaches that leverage known malware datasets. Customers will benefit from an additional layer of security that looks inside of individual files to identify malware and then automatically clears the file or blocks the spread of malware in near real-time.

These new capabilities provide customers using Box Shield with higher malware detection rates and fewer false positives. When dealing with billions of files, the benefits to both improved security posture and productivity are significant. Box Shield already scans over 48 billion files a year, greatly reducing content-centric risk and now with malware deep scan, Box Shield will be able to:

• Recognize malicious traits inside content in near real-time by leveraging the latest deep learning models to provide customers with broader coverage of sophisticated malware.
• Extend malware detection to active content in Box as users upload, update, download, preview, share, copy or move content to reduce the risk of malware infection by scanning both new and historical content.
• Analyze external content that is accessed by managed users to expand protection to content that is shared with an organization from an external source.
• Allow admins to occasionally override threat verdicts for low-risk content to avoid disrupting business workflows.

Box today also announced enhanced alerts powered by machine learning for anomalous user behavior like suspicious downloads in Box Shield. Admins will also receive more detailed alerts with context explaining why Box Shield’s machine learning algorithm has deemed certain behaviors as risky. These improvements will better equip admins in their investigation of anomalous behavior and provide granular feedback to train underlying algorithms for their company.

Marketing Technology News: MarTech Interview with Phil Gray, Chief Innovation Officer at Interactions LLC

Easily discover more sensitive content with auto-classification in Box Shield

In addition to helping detect and thwart potential insider threats or compromised accounts, Box Shield also uses advanced machine learning to help prevent accidental data leaks through a system of manual and automated security classifications for files, folders, and classification-based access policies. Auto-classification in Box Shield intelligently applies labels to files based on content inside, enabling customers to discover and label sensitive files at scale.

Box Shield helps customers reduce the risk of a data breach by automatically applying classification labels to 1.6 million files containing sensitive content like Personally Identifiable Information (PII). Earlier this year, Box expanded coverage beyond traditional regulated content to intellectual property and now supports more built-in info types like Canadian PII. To ensure files are protected as they move through workflows in Box, auto-classification now extends to active and existing content in Box. Additionally, Box released a deepened integration with Microsoft Information Protection (MIP) that enables customers across the two platforms to ensure that only authorized users get access to confidential data and that sensitive information is not shared unintentionally.

Marketing Technology News: Audience Targeting in Local Media Markets: More Than a Sum of Its Parts