The increase in the rate of high-profile threats (e.g., ransomware) reaches double-digit growth (15.8%). The result is a dangerous path that is likely to lead to continued losses for organizations victims of a cyberattack with no gain in defensive powers. In fact, a 2021 data breach report from IBM and the Ponemon Institute reveals that the average cost of a data breach is $ 4.24 million.
Beyond costs, a cyberattack can cause irreparable damage to a company’s brand, stock price, and day-to-day operations. According to a recent Deloitte survey, 32% of respondents cited operational disruption as the biggest impact of a cyber incident or breach. Other repercussions cited by surveyed companies include intellectual property theft (22%), falling stock prices (19%), loss of reputation (17%), and loss of customer confidence (17%). ).
Given these significant risks, organizations simply cannot afford to accept the status quo on the protection of digital assets. “If we ever want to overtake our opponents, the world has to change the mindset of detection to that of prevention,” Caspi says. “Organizations need to change the way they do security and fight hackers.”
Deep learning can be the difference
To date, many cybersecurity experts have seen machine learning as the most innovative approach to safeguarding digital assets. But deep learning is ideal for changing the way we prevent cybersecurity attacks. Any machine learning tool can be understood and theoretically reverse engineered to introduce a bias or vulnerability that will weaken your defenses against an attack. Bad actors can also use their own machine learning algorithms to contaminate a defensive solution with fake datasets.
Fortunately, deep learning addresses the limitations of machine learning by circumventing the need for highly skilled and experienced data scientists to manually feed a data set of solution. Rather, a deep learning model, developed specifically for cybersecurity, can absorb and process large volumes of raw data to fully train the system. These neural networks become autonomous, once trained, and do not require constant human intervention. This combination of a learning methodology based on raw data and larger data sets means that deep learning is able to accurately identify much more complex patterns than machine learning, at much faster speeds.
“Deep learning outweighs any denial-of-list, heuristic or standard-based machine learning approach,” says Mirel Sehic, vice president and CEO of Honeywell Building Technologies (HBT), a multinational corporation and supplier of aerospace materials. of performance and security and productivity. technologies. “The time it takes a deep learning-based approach to detect a specific threat is much faster than any of these elements combined.”
Download the full report.
This content was produced by Insights, the custom content group for the MIT Technology Review. It was not written by the editorial staff of the MIT Technology Review.