Proof Is In The Pudding For Deep Instinct

Marketing for any company or product attempts to spotlight its strengths while hiding or ignoring its weaknesses or issues. The company obviously wants you to do business with them or buy their products or services, so you need a healthy dose of skepticism and some critical thinking when evaluating their claims. Better yet—you need input from credible, independent, third-party sources.

The proof is in the proverbial pudding. The expression means that you can’t establish whether a pudding is good or bad based on how it is described. You have to actually eat the pudding for yourself. In other words, the value or quality of something can only be determined through actual experience.

Deep Instinct Product Assessment

Deep Instinct makes some bold claims about its deep learning AI (artificial intelligence) cybersecurity platform, but they don’t expect you to take their word for it. The company invited Unit 221B to thoroughly test and evaluate their threat prevention capabilities and validate—or refute—their claims.

Unit 221B conducted a thorough validation of the Deep Instinct platform. They tested it against portable, unknown, custom-designed attacks, and Python executables, and the efficacy of its static, dynamic, network, behavioral analysis, and signature detection.

The complete analysis and results are shared in the Deep Instinct Product Assessment report. Some of the key findings were shared in a Deep Instinct press release:

• 99.78% Accuracy – Deep Instinct exhibited a combined 99.78% accuracy rate for detection and prevention across unknown and custom attacks. Unit 221B tested Deep Instinct with a recommended configuration suitable for a mature customer’s hardened environment.
• 100% of Unknown Attacks – Deep Instinct successfully prevented 100% of unknown attacks and 96.4% of Unit 221B’s customized attacks.
• 60% Reduction – In events/alerts recorded to SIEM/EDR solutions with Deep Instinct installed compared to Microsoft Defender alone. This results in less strain on staffing and lowering of alert fatigue levels, allowing staff to be more focused on strategic and critical tasks such as patching and system hardening.

Perspective of a Professional Skeptic

I spoke with Lance James, founder and CEO of Unit 221B, about the assessment and his thoughts on the Deep Instinct platform. He explained that Deep Instinct believes in their product, so they wanted to put it to the test and prove to themselves that it lives up to their claims. They also want other people to believe in their product, and that is why they enlisted the support of Unit 221B.

“The best way to do that is obviously get together with some street cred security guys that will probably be harsh about the truth, no matter what,” declared James. “That’s where we come in.”

James told me he was relatively skeptical about the claims Deep Instinct makes when they began the assessment. He warned Deep Instinct that his team will not hold back—that they were going to get a comprehensive and honest assessment, and the truth might sting a little bit.

Prevention Is Better Than Detection

The mantra in cybersecurity over the past 10 or 15 years has been that you can’t prevent attacks. There has been a seismic shift from perimeter defense technologies to the concept of detect and respond. Products like EDR (endpoint detection and response) and XDR (extended detection and response) function on the premise that attackers will get through and exploits will happen—you just need to detect them faster and respond quickly before extensive damage occurs.

Deep Instinct rejects that mantra. Deep Instinct claims you can prevent attacks—even unknown or emerging attacks. Given the messaging of most of the vendors in the cybersecurity market, though, that is an understandably hard pill to swallow.

James pointed out that disruption requires being provocative. He added, though, that there are two ways of doing that: loud confidence or quiet competence. He stressed that loud confidence often comes with pushback or backlash, but quiet competence is taking a leap of faith and letting the results speak for themselves.

Ransomware Protection

One of the biggest cyber threats facing companies today is ransomware. If there is any threat where prevention is crucial and detection is often inadequate, it’s ransomware. Once an organization’s data has been encrypted with ransomware, there are no good options on the table.

James said, “It’s one of the biggest threats, especially small to medium businesses—they just don’t have the tolerance. I’ve watched medium-sized businesses literally during COVID have to go under, and that was painful.”

According to Unit 221B, Deep Instinct lives up to its claims when it comes to preventing ransomware. James told me that the Unit 221B tried REvil, NotPetya, and other notorious ransomware exploits, and they also developed custom ransomware of their own using new techniques. Deep Instinct stopped them all.

He added that they also tried a standard, commercial encryption tool, and Deep Instinct was smart enough to recognize that it was a legitimate application and not flag it as a false positive.

James emphasized, “Even if you bought Deep Instinct just for ransomware, I think you’d sleep better at night. If there’s anything I want to do prevention of, it’s ransomware.”

Validating Deep Learning

“Customers select Deep Instinct for our unique, pre-execution approach to stopping zero-day attacks,” proclaimed Guy Caspi, CEO of Deep Instinct in the press release for the Unit 221B report. “While most organizations focus on detection, we aim to prevent an attack before it gains access to your organizations’ infrastructure because, in many cases, detection is too little too late. I’m grateful for the team at Unit 221B for working on this evaluation.”

You can download the full results and details of Unit 22B’s product assessment of the Deep Instinct Prevention Platform, and see for yourself: Deep Instinct Product Assessment.