• 07/08/2025
  • Industry News

How security specialists conduct cybersecurity research

Security companies around the world are conducting research in internal laboratories to discover security vulnerabilities. How do they go about it? A look behind the scenes shows their technology and methods - and what they find.

Written by Uwe Sievers

Medical monitors with warnings, Windows XP and clock displays on the shelf.

Security gaps in routers become the target of hackers

Modern IT and OT devices are not always free of security problems. To prevent these from being discovered by a cyberattack, security companies do everything they can to detect security vulnerabilities at an early stage. Quite a few have set up their own lab for this purpose and analyse a wide range of devices under laboratory conditions. Very few, however, allow an insight into their facilities. We were there.


If an IT problem cannot be solved by a restart or an update, it is really serious. A number of Asus routers are currently affected. Attackers have infiltrated these devices en masse and installed a backdoor that grants them permanent access. They hacked the devices using simple brute force methods, i.e. mass attempts to enter login data. They firmly anchored the backdoor in the NVRAM, which means that it cannot be easily removed by restarting the device or switching it off. The only way to remove this backdoor is to reset it to the factory settings. However, the complete configuration is lost and must be carried out again.

Only recently, a study by security specialist Forescout showed that routers are among the most risky devices in company networks. The study revealed an interesting change in this respect compared to the studies of previous years: network components, especially routers, are more popular targets than classic IT components such as PCs and servers. "Routers account for over 50 per cent of devices with critical vulnerabilities," the study states. According to the study, network infrastructure components are more at risk than end devices. The risk factor depends on their position at the interface of networks, usually between internal and external networks. The more network areas they connect, the more interesting they are for attackers, because the more data traffic passes through them. From there, it is usually quite easy to manoeuvre through the company's network architecture. According to the study, this in turn is made easier by the fact that these devices often have open or poorly secured administrator accounts.

Research results shape studies

The study was largely the result of research conducted by the US company in its security laboratory in the Netherlands, known as the Vedere Lab. The lab, which specialises in Operational Technology (OT), is located in Eindhoven in the immediate vicinity of the university. The proximity to science ensures an up-to-date level of knowledge and new specialists. Daniel dos Santos explains that it is possible to intensively analyse devices that are used in industry and business here. He has a PhD in computer science and heads the laboratory. It is not only used for testing and research, but also for demonstration purposes, for example for customers. The lab has repeatedly won prizes for its research, including this year.

In addition to industrial components, other OT and IOT systems are also part of the core area, including medical devices and building management systems. There are a large number of different systems in each area. The collection of devices in the lab is impressive. It took over ten years to set up the lab, procure the most important devices and analyse them, explains dos Santos. The most common industrial components are neatly lined up on top-hat rails in the lab, sorted by function and type. At the centre are Industrial Control Systems (ICS), including classic SCADA elements or components of programmable logic controllers (PLCs), which can be found everywhere in automated production.

 

No shortage of security problems

There are also typical devices from hospitals and other medical facilities. For example, an image management system which, among other things, enables images and video recordings of various formats to be taken and displayed in operating theatres or treatment rooms. However, it still runs on Windows XP and is therefore quite easy to attack. The lab is also home to newer devices such as a multi-parameter analyser for monitoring blood sugar levels in diabetics and for the early detection of kidney disease. In other words, devices that can be vital for patients.

They are all available for examinations and demonstrations. However, hardly any of them are free of security problems, according to voices in the lab. Devices often have default passwords. These are sometimes included in the operating manual or can even be found on the manufacturer's website. These are extremely easy attack scenarios for attackers to realise. There are even search engines to find such devices. They can also search specifically for the locations of components with certain security vulnerabilities. However, not only attackers benefit from this, but also the other side, i.e. security authorities or companies. For example, they can warn operators of emerging problems at an early stage. "We also analyse anonymised customer data for this purpose," explains Christina Höfer, Head of OT and IoT Strategy at Forescout. This allows us to know "where, for what and how often these devices are used", she adds. These results are also used in the lab. New studies are regularly produced as a result.

 

Intentional hazard

When you enter the lab, you notice a particularly secure and separate room. The security specialists call it the "Danger Zone", as the systems in it are exposed to the Internet relatively unprotected. This is the only way to observe the behaviour of attackers and study their methods, explains an employee. Among other things, a complete industrial environment consisting of several virtual machines is simulated here on a virtualisation server. It is supplemented by cloud systems. Routers, edge devices and engineering workstations are also included. Everything is completed as required by other OT or IT systems to be analysed. The entire ensemble serves as a playground for attackers and is known in IT as a honeypot. Every action and every step taken by an intruder can be observed in real time; you can look over the shoulder of an attacker live, so to speak. This allows new attack variants and approaches to be analysed very quickly. "For an attacker, it looks like a real production plant or a hospital, depending on the configuration," explains dos Santos. And it can be adapted very flexibly and quickly. All this to stay one step ahead of the attackers. However, they have long been aware of honeypots and usually check for evidence of such traps before launching attacks. It is therefore a challenging endeavour to design honeypots in such a way that they do not stand out as such.

The experts in the lab were also able to discover a new trend: the increase in indirect attacks. These are carried out via building management and automation systems, for example, which are often inadequately secured. Attacks on power supply components are therefore conceivable. If these succeed, a production facility or hospital could be severely disrupted. If security companies can use their research laboratories to help prevent such attacks, they can sometimes help save lives.

 

 


Wie sich Unternehmen schützen können, zeigt unser Überblick zum Thema OT-Security.

Erfahren Sie auf unserer Themenseite „Cyberangriffe“, wie Sie Ihr Unternehmen ganzheitlich vor Cyberangriffen schützen – mit Hintergrundwissen, Best Practices und aktuellen Trends.