Andrew Walenstein is Director of Security Research and Development for BlackBerry Cylance. The first computer virus. Many people don't know that the first computer virus predates the Internet. In 1971, Bob Thomas, an American computer science scholar, wrote Creeper, the first computer program capable of network migration. He would travel between the ARPANET terminals printing the message "I am the climbing plant, catch me if you can". Creeper was created by fellow academic and email inventor Ray Thomlinson to create the first documented computer virus. To contain the Creeper, Thomlinson wrote Reaper, a program designed to kick the Creeper off the network and delete it, creating the world's first anti-virus cybersecurity solution. How cybersecurity has developed. At that time, it would have been hard to imagine that a virus as simple and harmless as Creeper could be the forerunner for the development of malicious software and destructive ransomware such as ILOVEYOU and WannaCry. Fortunately, modern cybersecurity has come a long way since Reaper. Today, any mention of cybersecurity will inevitably lead to discussions of security solutions based on artificial intelligence (AI) and machine learning (AI). In fact, the next generation of cybersecurity threats requires agile and intelligent programs that can quickly adapt to new and unexpected attacks. The potential for Amnesty International and ML to meet this challenge has not escaped the attention of cybersecurity policy makers, the vast majority of whom believe that AI is critical to the future of cybersecurity. Despite the hype, however, many decision makers still don't know exactly how AI and ML-based security products work.
AI and Cybersecurity Recently, "neural network" artificial intelligence techniques have become extremely popular and have led one to think that they are shiny and new. However, many are often surprised to learn that AI is not a new phenomenon. Artificial intelligence is not the latest, neural networks have been around for over half a century, and some of the first commercial neural networks for malware detection and destruction were developed over 20 years ago. Years: Protecting Boot Sectors from Floppy Disk Viruses in the Windows 98 Era. Machine Learning Techniques. Another thing that seems surprising is the number of places where ML is involved in protecting systems. This could be due to people's reactions to the "machine" part of ML. Actually, BC is just another way of learning from examples, a concept that everyone can understand. So, whether it is a human or a machine learning to perform a task, the only thing that matters is the level of sophistication and the know-how that is gained. A good example is the predictive keyboard on your smartphone. It contains a small machine-learning engine that reads what you type and draws lessons from your writing style to predict what you might say next, or at least what you intend to say. As you feed the text, he can learn more confidently and accurately what he personally says and how he says it. The value is that you have your own non-human assistant capable of predicting your speech. Instead of a predictive keyboard, if we feed ML your typing, mouse, and other activities, it can learn even more about your unique behavior, becoming adept at recognizing you and your little idiosyncrasies. Instead of entering text, if you enter it into malicious software, you have a malware detector. Feed network attacks and you have an IDS. These and many variants are found in network and PPE products. This is the first type of cybersecurity application that many people think of AI, and it is probably the most prevalent and the most mature. In practice, machine learning is much more complex than simply asking a computer to solve a problem. As for Creeper and Reaper, the development of ML and AI-based threat detection requires a high degree of understanding, based on experience, as well as an innovative approach that always leaves a little behind. ahead of the attackers. Andrew Walenstein is Director of Security Research and Development for BlackBerry Cylance.