Deep Instinct Unleashes DSX: A Revolutionary Leap in Cyber Defense

Deep Instinct Unleashes DSX: A Revolutionary Leap in Cyber Defense

Summary

• Deep Instinct has introduced DSX, a security platform that tries to stop threats before they run.
• DSX uses deep learning to judge files it has never seen, rather than matching them to known signatures.
• Industry recognition: DSX has been nominated for the Cubed Awards.
• Who built it: Deep Instinct’s team has applied deep learning to security since the company was founded.
• DSX aims to prevent attacks rather than detect and clean up after them.

Deep Instinct’s DSX bets on prevention

Deep Instinct, one of the first companies to apply deep learning to cybersecurity, has unveiled DSX. The pitch is prevention. Most security tools alert after something malicious starts running. DSX tries to block the file before it executes.

How the deep learning model works

DSX runs a deep learning model trained on large volumes of malicious and benign files. The model learns structural traits of malware, so it can judge a file it has never seen instead of looking it up in a list of known threats. That matters for zero-day malware, which by definition has no signature yet. Deep Instinct co-founder Guy Caspi has long argued that detection often comes too late and that stopping an attack before it gets in is the real problem to solve.

The team and the award nomination

Deep Instinct was founded in 2015 by Guy Caspi, Nadav Maman and Eli David, and its staff includes machine learning researchers alongside security engineers. DSX has been nominated for the Cubed Awards, an industry recognition program.

Prevention instead of cleanup

Traditional security stacks detect, alert and then respond, which leaves a window between the first malicious action and containment. A model that blocks files pre-execution aims to close that window. If it works as advertised, analysts spend less time cleaning up infections that started before anyone noticed.

What buyers should test

DSX is one more sign that security vendors are moving from detection toward prevention. Buyers should test the claims on their own data: false-positive rates and coverage of the file types they actually store matter more than launch language.

For teams drowning in alerts, a tool that blocks more threats up front would be welcome, if it holds up in practice.

Frank Jones, CISSP

Frank Jones has loved computers from the age of 13. Frank got his hacking career started when he downloaded a war dialing program that he used to detect dial up modems in his hometown of Chicago. Frank Jones now works as a JAVA coder and cyber security researcher.