Deep Instinct has developed the first deep learning cybersecurity framework, and studies show it has a 100 percent threat detection rate.
Tech Data has operations in over 100 countries and in March it reported US$37.5 billion in net sales for the 2019 fiscal year. As part of the collaboration, Tech Data will offer clients in Latin America, Canada and the US Deep Instinct’s cybersecurity framework.
Deep Instinct has developed a deep learning framework designed specifically for cybersecurity, which it says is the first framework of its kind.
“Our customers are always looking for the best, most advanced solutions to support their security needs, and, based on what we have seen from Deep Instinct, our customers are going to appreciate the value this solution will bring to their organizations,” said Yuda Saydun, president of CyVent, Tech Data’s channel partner, in a press release.
For Deep Instinct, the partnership will allow it to leverage the scale and relationships of Tech Data to further extend its deep learning product.
“Deep learning is much different than what’s out there in the market,” Grady Johnston, vice president of channel and alliances at Deep Instinct, said via phone. “A lot of it is machine learning, where you still need people in the background telling the application what to look for and what to change.”
Instead, Deep Instinct feeds computers files, which the computers learn to identify on their own.
“With deep learning, it’s about what we call a brain, where we actually create what we call our own little university with multiple huge computers,” Johnston explained. “We feed it tons of millions and billions of files of good and bad, and this brain learns what’s good and bad by the bytes.”
Results from SE Labs have shown 100 percent threat detection and prevention rates from Deep Instinct’s framework for identifying malware and viruses.
“So no matter what attack that comes or has been mutated, our brain will see it because it looks at every little byte, so it stops it before execution. We are a prevention company,” said Johnston.
Where deep learning is distinguished from machine learning is that machine learning requires feature extraction, which is when humans tell the computer what different characteristics or properties are.
Take, for example, classifying spam emails. Specific features are tied to different types of emails and through machine learning, the computer learns to automatically file spam emails into a separate folder. The same process has been applied to image recognition, voice recognition and diagnosis.
Deep learning takes this a step further. As a subset of machine learning, deep learning processes data without the need for feature extraction.
According to Eli David, chief technical officer and co-founder of Deep Instinct, deep learning has illustrated the highest level of performance improvement in the history of artificial intelligence (AI) in areas including text and speech recognition, as well as computer vision.
Currently, gaps exist within enterprise cybersecurity solutions. The recent WhatsApp and Norsk Hydro (OTCQX:NHYDY,OL:NHY) attacks, as well as the 800 other reported data breaches reported in 2018, shine light on the current security atmosphere.
Cutting down on false positives and improving prevention methods is beneficial for companies. “We’ve got to really start looking at leveraging new ways of doing things, such as the AI deep learning that’s used in smart cars and a lot of other stuff. Technology has to go to the next level, so we are the first ones there,” said Johnston.
Looking forward in the cybersecurity sector, while there are a number of inefficiencies that continue to be targeted, deep learning offers one alternative prevention method. Take, for example, single sign-on products. Because single sign-on products are built on open-source code, hackers can almost immediately inject malware and steal this information. They can then log a user’s account.
At this point in time, hackers are continually using machine learning as part of their attacks, and the level of sophistication continues to grow. When it comes to prevention methods, deep learning has the ability to take a more comprehensive analysis of what is happening. Instead of looking at specific files, like machine learning does, it has the ability to look at all of them.
These capabilities then help prevent any unwanted outcomes, leveraging the knowledge and scale that deep learning has the ability to provide.
Don’t forget to follow us @INN_Technology for real-time news updates!
Securities Disclosure: I, Dorothy Neufeld, hold no direct investment interest in any company mentioned in this article.
Editorial Disclosure: The Investing News Network does not guarantee the accuracy or thoroughness of the information reported in the interviews it conducts. The opinions expressed in these interviews do not reflect the opinions of the Investing News Network and do not constitute investment advice. All readers are encouraged to perform their own due diligence.