AI can protect legacy devices from cyber attacks

2020-01-01

In this era of informatization, almost every company is vulnerable to cyber attacks. What's worse, many companies are unsure of how to prepare and respond to such attacks. The larger the company and its devices, the greater the challenges. Protecting an outdated device is much more complex than downloading the latest antivirus software, and most managers don't know where to start. According to a survey conducted by the Bonn Research Institute for Cyber Security in the power industry, more than half of the executives expect a critical infrastructure to be targeted by cyber attacks in the next 12 months, and they are not confident in their company's defense capabilities. The main issues faced by these companies include a lack of highly skilled programmers, fragmented control systems, and slow discovery of system vulnerabilities. This is not an isolated case, as other industries that rely on large computer-controlled machinery, such as oil and gas, refining, and manufacturing, also rank low in terms of cybersecurity. The main reason for this problem is that unlike software that is replaced every two to three years, their devices are designed to have a lifespan of 20 years or longer. Leo Simonovich, Global Head of Industrial Networks and Digital Security at Siemens, said, "On the one hand, there are a large number of brownfield assets that have never been considered for security during their design. On the other hand, the energy sector is undergoing a significant transformation through digitization and interconnectivity with the introduction of renewable energy. Sometimes, patching network vulnerabilities is worse than the risk of attacks. If you perform a basic vulnerability scan, you can destroy a factory. If you deploy a patch, you can even destroy a major corporation or a fleet. For various reasons, 90% of things cannot be patched; they are either remote or would disrupt the production environment." In a world where remote operations and cloud computing have become the norm, one option is to disconnect systems from the rest of the world, creating what is known as an "air gap." Another option is to establish detection systems that enable control engineers to quickly identify and confidently prevent network attacks. Simonovich said, "Understand what you have, how vulnerable it is, and then prioritize based on business priorities, operational priorities, and risk priorities, and take practical action. This is where artificial intelligence can truly come into play." Trained artificial intelligence can detect abnormal activities in specific technologies much faster than humans. With a database created from past experiences, it can recommend the right course of action. However, it is important to note that this does not mean complete peace of mind or a one-time solution. Humans still need to be fully prepared and address any vulnerabilities and gaps.

RECENT AI NEWS

RECENT AI TOOLS