It is with pleasure that today we are talking about Deep Instinct, a solution that uses deep learning as a tool to prevent, anticipate and counter potential new malicious threats.
We asked some questions to the Deep Instinct team who told us how the cyber security landscape is evolving, always looking for new products that are increasingly targeted and performing.
1) What are the urgent needs of Cyber Security today?
As organizations are migrating and relying heavily on the cloud to conduct their business and maintain productivity with remote workforces, the need to protect the cloud is greater than ever. Millions of employees are now using their personal smartphones, tablets and laptops for work – and these devices may be easier to compromise, which directly affects organizations when these devices access protected areas such as data storage, sensitive applications, and critical environments from those unprotected and unmanaged devices. This puts a massive strain on already limited security resources in many organizations leading to alert fatigue and burn out which can ultimately lead to missing critical security alerts and incidents.
2) How can deep learning help the identification of threats compared to other A.I. solutions?
The biggest problem in security was and still is prevention. Deep Instinct is the first deep learning cyber security framework that was purpose built for the prevention of cyber-attacks. Deep Learning technology and its ability to constantly adapt and protect against the ever-changing threat landscape allows us to do stop unknow, never seen before threats in less than 20 milliseconds with a false positive rate of less than 0.1%, which is the lowest in the industry.
Inspired by the human brain’s ability to learn, deep learning models develop the innate ability to accurately distinguish malicious files and processes from benign ones, in milliseconds. As a result, any kind of threat, known or unknown, whether first-seen malware, zero-day threats, ransomware, or APT attacks of any kind are predicted and prevented before they can execute.
3) What is a typical attack on which Deep Instinct reacts predictively?
All kinds of conceivable threats and attacks (malware, unknown, known, zero days, fileless attack). We are continuous feeding our deep learning “brain” on 100’s of millions of new malicious samples, code, and attacks ensure our predictive deep learning technology in the forefront of the newest emerging threats.
4) How does the threat detection and resolution process work?
Deep Instinct’s artificial deep neural network brain learns to prevent any type of cyber threat, its prediction capabilities become instinctive. As a result, any kind of threat, known or new, zero-day (first-seen malware), ransomware, and APT attacks are predicted and prevented before it can execute, effectively in zero-time. Unlike detection and response-based solutions, which wait for the execution of an attack to react, or post analysis which achieves too little too late, our prevention approach keeps customers protected, while dramatically reducing their costs.
5) What is the interaction with the user and this solution?
No action from users is required. It works independently, even when the system is offline and not connected to the internet.
6) 3 strengths that describe Deepinstinct?
• We do not use Virus Pattern/IOC/IOA Base. We work without the critical need to update the solution or to be connected to internet.
• Effective against unknown, first seen attacks (APTs, Zero-days etc.)
• We block the threats pre-execution while in the cache of the system before they access to the disk. We are the only solution to do this. Our patented Deep learning speed allows it.
7) What could be the evolution of a system like Deep Instinct be in the future of cyber security scenario?
Cybersecurity is the top spending priority for CIOs globally, it has been for the last 20 years and will be for the inevitable future. The attack vectors, sophistication of the attacks, and the sheer volume of attacks continues to grow exponentially and has only been exacerbated by working from anywhere. As we are the leading deep learning cybersecurity company, protecting not only endpoints but also the cloud and applications by leveraging the same deep learning platform, our vision of the next phase of protection is prevention before the endpoint.
8) 3 Deep Instinct tips for users to strengthen their defenses?
• Use of a purposed build deep learning framework as a predictive and preventive technology against todays advanced cyber threats.
• The use of a layered security stack where security tools augment and firm up each other. For example, complementing the stack with an installed EDR/XDR solution to identify where threats are trying to take hold in our environment. Just because a solution like Deep instinct can prevent the attacks, you still must find and fill then security holes in the infrastructure of where these attacks are entering the environment. This limits the exposure and scaled down the risks significantly.
• Customer’s security teams need to spend less time reactivity chasing false positive and more time proactively on threat hunting. This will allow them to adhere to basic good security hygiene having time to keep their existing critical investments patched, maintained, monitored, and performing effectively and securely.
9) What is the difference between ML and DL applied to cyber security solutions?
One of the main differences between DL and ML is the Predictive ability; Deep learning leverages deep neural networks that can solve tasks that machine learning models can’t because Machine learning requires a human domain expert to supervise, define, and engineer features for the machine to learn. Deep Learning learns from millions of files of raw data without any supervision or interference from humans. The result, quite simply, is that deep learning is far more accurate than machine learning based approaches. There is no manual feature engineering, so it’s far harder for malware to understand how we work and then to overcome and bypass our solution. Despite all this huge computing power behind the Deep Instinct solution, the actual footprint required is tiny – you truly have the best of both worlds.
