“Deep Learning enables us to train our brain on 100% of raw data which enables us to make more accurate decisions, faster than ML-based solution that can only train on about 2-5% off available date.”
Hi Guy, welcome to the iTechnology Interview Series. Please tell us about your journey in the cybersecurity industry?
After years in Israel’s Defense Forces’ cyber elite team, I used my technological background to move to the private sector and joined Comverse, which later became Verint (VRNT/CGNT) & Mavenir (MVNR). For a decade, I held c-level executive positions in the Nasdaq traded companies, which included product development and commercial/sales. In 2015, I along with Eli David and Nadav Maman, co-founded Deep Instinct. Dissatisfied with the available cybersecurity companies, we dedicated ourselves to pursuing the uncharted territory of deep learning to find a security product that would stop even the most sophisticated types of attacks. Together, we founded the world’s first and only deep learning cybersecurity company.
What is Deep Instinct? What inspired you to start Deep Instinct?
The Deep Instinct Prevention Platform prevents unknown threats faster, with greater accuracy and lower false positives than any product on the market. We can predict, prevent, and analyze cyberattacks at any touchpoint of the organization from the endpoint through to the network and in the cloud. Deep Instinct is pioneering the deep learning adoption in cybersecurity, much like autonomous vehicles, speech recognition and recommendation engines have leveraged deep learning to turbocharge their applications.
Deep Instinct’s cybersecurity platform utilizes end-to-end deep learning to specialize in threat prevention, making it the most efficient and effective cybersecurity solution in the market. We have developed the only deep learning framework purpose-built for cybersecurity, allowing Deep Instinct to prevent threats that have never been seen before and do it faster than other endpoint solutions.
I was on Israel’s Defense Forces’ cyber elite team, which plays a critical role in national security. From that experience, I know that what is happening today in terms of threat actors’ sophistication and complexity is out of control and will only continue. What most people see or read is just the tip of the iceberg. The status quo approach to security is antiquated, reactive, and insufficient.
The biggest problem in security was prevention and still is. While most organizations focus on detection, we aim to prevent an attack before it gains access to your organizations’ infrastructure because, in most cases, detection is too little too late. We sought to solve the pervasive industry issues of telemtry overload by developing the industry’s only innovative deep learning solution that prevents first and then sends the data for analysis not the other way around.
How did your role evolve through the pandemic? What unique challenges did you tackle in the last 2 years that you think was key to Deep Instinct’s growth in the cyber intelligence domain?
As businesses are relying 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. With the massive shift to home offices, we’ve had to increase scale to meet the security needs of our customers. My role became more dynamic. With the absence of face-to-face meetings (offices closed; no travel, etc.), the need to communicate effectively with all of our stakeholders became much more critical. Also, per McKinsey’s The CEO moment: Leadership for a new era leadership isn’t just about business decisions, but about supporting and elevating team members’ morale and inspiring a good workplace culture. The entire leadership team increased our commitment to and investment in our overall employee experience.
Which markets are you currently targeting? What does your ideal customer profile look like in the cyber threats management landscape?
Any enterprise organization who is looking to prevent ransomware, zero days and other unknown threats before they reach the endpoint, servers or storage to improve their overall security posture will benefit from Deep Instinct. In particular financial services organizations, retail, manufacturing, and healthcare are counted amongst our customers. We are also an important technology for MSSP, and other critical infrastructure environments.
Tell us a little bit about your IT infrastructure and how it helps you build a great product at Deep Instinct?
Our IT infrastructure is optimized to produce, test, and validate the newest and most sophisticated deep learning models to prevent today’s most advanced threats facing organizations throughout the globe. In our lab we rely heavily on Nvidia GPUs due to their inherent memory bandwidth architecture allowing us to handle huge amounts of data to build our neural networks. GPUs greatly enhanced the data science models computations when they are carried out in parallel, which is why is what GPUs are designed for.
How do you use Deep Learning techniques to improve cyber threat management/ ransomware detection?
Our deep learning frameworks are purpose built from the ground up to prevent cyber security threats. This means that Deep Instinct is able to “predict” malicious threats that have never been seen in the wild before they are even coded. Our neural networks mimic the human brain, scan in the entire file, and make a malicious vs benign determination in less than 20 milliseconds, 750x faster than ransomware can encrypt.
Deep Learning enables us to train our brain on 100% of raw data which enables us to make more accurate decisions, faster than ML-based solution that can only train on about 2-5% off available date. Deep Instinct has lowered the false positive rate to an industry leading 0.1% and the is because our platform is built to take advantage of deep learning.
Overall Deep Instinct is able to prevent ransomware flat out before it can get inside meaning it never gets the chance to exfiltrate or encrypt data. Another impact is that we significantly reduce alert fatigue of SOC teams and provide them with time back to do more meaningful and important work.
Other security solutions today depend on the file to be executed and then they look for patterns or markers post execution looking for malicious type behavior. Our deep learning platform does not require this. We prevent the malicious attacks before they execute.
Read More: ITechnology Interview with Jeremiah Stone, CTO at SnapLogic
We are curious to learn more about your Deep Instinct Prevention for Applications offering, Could you please provide a sneak preview of what it does and how it makes end-to-end cyber security management a real possibility?
As zero-day, ransomware, file-based, fileless, supply chain attacks and new malware variants continue to morph and evade detection, it has become exceedingly difficult to stop bad actors from succeeding once they get inside a hybrid network. Deep Instinct is reversing this trend to prevent unknown attacks before they can land in an environment through endpoints, servers, or storage by leveraging the power of deep learning.
At the core of the Deep Instinct Prevention Platform is our industry-leading antimalware static analysis built to take full advantage of the power of deep learning to make faster, more accurate decisions and prevent threats before they can fully execute and land in an environment. Deep Instinct Prevention for Endpoints provides additional layers of prevention with dynamic and behavioral analysis to prevent files attacks, like in-memory and code injection, and better protect against ML evasion and poisoning techniques. Deep Instinct Prevention for Applications is focused on preventing malicious files from impacting network and storage environments.
How have new technologies such as AI, ML, DL, Cloud computing and analytics transformed the global IT security and data ops services industry?
AI has been used in the IT security space for greater than 10 years. This started as very basic with heuristics, then grew into utilizing machine learning for pattern recognition, and now deep learning neural networks. Historically machine learning has been at the backbone of today’s common security products, but the issue is they are dependent on files to be executed, and are then examined for behaviors, techniques and tactics are determined to be malicious or benign. This is a challenge for today’s security tools utilizing machine learning, because the speed at which a threat actor can move laterally within an environment and infects systems is faster than can be detected and responded to before any damage is done. Also, threat actors are continuing to evolve threats to evade EDR, sandbox and AV tools in addition to using ML and adversarial AI attacks that now in the hands of common cyber criminals.
Advances in deep learning and purpose-built solutions have been specifically designed and architected to address these shortcomings. Deep instinct has a proven track record with its deep learning solution and has proven to be immune to these types of adversarial attacks due to its purpose-built neural networks.
Talent War and Talent Shortage have slowed down the pace of many industries. How do you tactically overcome the challenges arising out of talent management or mismanagement in your industry?
Hiring and retaining talent is a long-term investment. Just as companies want their employees to be passionate and dedicated to their job, they must reciprocate with smart, common-sense approaches to making job environments healthy and rewarding. Organizations must help employees to do succeed in their job by providing the best tools on the market that will optimize their performance. They must also take a holistic approach to the hiring process, onboarding, remote working environment with continuous development and a superior overall employee experience. With a focus on best-in-class technology and working environment companies can recruit and retain the cybersecurity talent they need to thrive.
How do you address talent shortages?
Savvy employers entice new talent and retain important security professionals by focusing on progressive employee experiences. In an ever-changing and evolving work environment, we are fully committed to supporting our workforce each step of the way, from the onboarding process to training and career development, to providing individual experiences that enhance the day-to-day for our employees.
Given that offices are no longer the hub of the working day, we invest in ways to give employees the same supportive and collaborative experience when most workers will operate from home part-time. We focus on creative ways to support a changing work environment by staying competitive in today’s fast-evolving work culture.
Data privacy and cyber threats continue to harm organizations with unprecedented, and most often, with irreversible outcomes. How do you ensure protection from the invisible cyber threats?
The devastation that sophisticated cyber threats are causing to organizations is catastrophic. Deep Instinct was built with the goal of addressing this issue head on. Today’s cyber threat landscape can be described as evolving from a cat-and-mouse game to a multilevel space-race with each side getting more powerful and devastating weapons. The problem is the malicious threat actors have been winning the war and the defenders are constantly in a reactive state. By employing a new approach with a deep learning cyber prevention solution, the defenders can move into a proactive state. They can prevent the attacks from even happening, giving them more time to secure their environments and educate their end users.
Read More: ITechnology Interview with Arti Raman, CEO at Titaniam
Your predictions on the future of data analytics and how most industries could benefit from adopting this technology to meet global IT storage and data security demands:
Deep learning is enabling the most advanced cyber defense available today. But you can’t apply just any deep learning framework to cybersecurity. It needs to be developed and trained specifically to solve the unique challenges that cybersecurity presents. Organizations need to continue adapting and utilizing solutions like those that Deep Instinct has architected. Industries need to stop waiting for attacks to happen and then react to them, they need to go on active offense and adopt a prevention first strategy.
Your advice to business leaders in IT and cybersecurity looking to adopt to ever-expanding data analytics and AI technology innovations for digital transformation
We need to get out ahead of the threat actors. Stop being complacent with your legacy security solutions and waiting for an attack to happen. Evaluate and adapt new technology as your adversaries are not standing still and are continually adapting to evade controls and develop new forms of malware faster than you are catching them.
Technology event/ conference/ webinar/ podcast that you would like to attend in 2022-2023:
A highlight every year for me is our presence at Black Hat. In May, we had the honor of speaking at Black Hat Asia in Singapore, and this week we will be on stage at the US event.
Thank you, Guy! That was fun and we hope to see you back on itechnologyseries.com soon.
[To participate in our interview series, please write to us at sghosh@martechseries.com]
Mr. Caspi is a seasoned CEO and serial entrepreneur with more than two decades of enterprise technology experience in executive C-suite positions. His vision and leadership have helped guide organizations to meet growth and revenue targets in several stages of maturity, from early startup through post-IPO. Caspi is a thought leader with deep domain expertise in deep learning and AI assimilation in cybersecurity and was selected as a Technology Pioneer by the World Economic Forum in Davos, Switzerland, in 2017. Additionally, he was nominated as a board member of the prestigious Forbes Technology Council in 2018. He holds a Master’s degree from Hebrew University in Jerusalem, and completed a Global Executive Program from the joint MIT Sloan Business School / Israeli Technion Institute.
Deep Instinct takes a prevention-first approach to stopping ransomware and other malware using the world’s first and only purpose-built, deep learning cybersecurity framework. We predict and prevent known, unknown, and zero-day threats in <20 milliseconds, 750X faster than the fastest ransomware can encrypt. Deep Instinct has >99% zero-day accuracy and promises a <0.1% false positive rate. The Deep Instinct Prevention Platform is an essential addition to every security stack—providing complete, multi-layered protection against threats across hybrid environments.