The Growing Challenge of Securing IoT Devices

The Internet of Things (IoT) continues to push the boundaries of convenience, allowing us to remotely monitor objects, automatically adjust the thermostat, or play music without tangled cables.

While we enjoy the convenience of these devices’ exponential growth, those responsible for securing them are having difficulty managing them or even locating them on the network.

This is despite the fact that new IP addresses under IPv6 provide internet addresses for the trillions of IoT devices that we expect to be connected to networks in the coming years.

What started as a network of relatively simple gadgets has grown into an ecosystem spanning billions of devices worldwide.

For enterprises, the simplified design and complex issues associated with medical cameras and monitors pose a threat, requiring urgent attention and innovative solutions.

Vulnerable by Design: Inherent Vulnerabilities in IoT Devices

One of the main problems with IoT devices is their inherent lack of robust cybersecurity features. Many of these devices are designed to perform specific functions without direct human interaction, such as a sensor monitoring temperature or humidity in a manufacturing facility.

Because many of these devices are seen as “set it and forget it,” they don’t have a user interface—they simply collect data and send it to a central collection point. However, this simplicity often means that these devices lack advanced security measures, leaving them vulnerable to attack.

This doesn’t even address the problem of identifying and managing IoT devices on the network. Unlike traditional computing devices, IoT devices may not be easily visible or identifiable in an organization’s IT infrastructure.

It’s nearly impossible to implement effective security measures without knowing what devices are present on the network. This visibility problem is exacerbated by the sheer number of devices, often called “shadow devices,” that can connect to the network without proper oversight.

The first major IoT breach sent shockwaves through the security and automotive industries. In July 2015, security researchers Charlie Miller and Chris Valasek exploited a zero-day flaw in a Jeep Cherokee’s entertainment system to disable features like braking, steering, and acceleration.

A Wired reporter was a “digital crash test dummy” and documented that “hackers” were able to connect to a Jeep wirelessly over the internet and take full control of the vehicle.

Another notable incident was the Mirai botnet attack in 2016, where a massive number of IoT devices were compromised and used to launch a distributed denial-of-service (DDoS) attack, which resulted in the system being overwhelmed with access requests to the point that it was unable to handle the load and crashed.

This attack exploited weak security features of IoT devices, such as default passwords, highlighting the urgent need for improved security practices and tools to defend against such threats.

Unfortunately, the number of such attacks has been steadily increasing, increasing by 108% year-over-year in the first quarter of 2024. Worse still, DDoS attacks in 2023 often targeted almost 4,000 devices, while in January-March this year, the number increased by 400% to over 16,000.

Strategies for Robust IoT Security

Given the scale and complexity of IoT networks, traditional security measures are often insufficient. This is where AI-based solutions come in.

AI can provide the scalability and adaptability needed to manage and secure the ever-increasing number of IoT devices in the field by continuously monitoring network traffic, identifying anomalies, and responding to potential threats in real time.

Such 24/7 monitoring is something that the Central Security Offices could only dream of. It is a dynamic defense mechanism against cyberattacks.

Best practices for securing IoT devices:

• Supplies management– Organizations need to maintain an accurate and up-to-date inventory of all IoT devices connected to their networks. This requires tools that automatically discover and catalog these devices, providing a clear picture of the network landscape.
• Default Password Policy– Many IoT devices have default passwords that users often do not change. Ensuring that all devices have unique, strong passwords is key to securing them.
• Network segmentation– By segmenting IoT devices into isolated networks, forcing them into digital “containment,” organizations can limit the potential damage caused by a compromised device. This approach ensures that even if one device is compromised, an attacker cannot easily move laterally across the entire network.
• Behavioral monitoring– Implementing continuous behavior monitoring can help detect suspicious activity that may indicate an infected device. AI-based tools can analyze patterns and flag deviations that could indicate an attack.

The Way Forward: Implementing a Comprehensive IoT Security Strategy

The reality is that IoT devices are here to stay — and for good reason. They make it easier to perform tasks that would otherwise bore humans, limit the effectiveness of security measures, or unnecessarily drain resources. They also help turn incidents into streams of data that can then be broken down and analyzed.

Given these realities, security teams should prioritize effectively protecting IoT devices and the networks they connect to. Fortunately, AI can use layman’s terms to help those looking to secure their operations by:

• Discovery and diagnosis– Understanding what devices are present and what their current security posture is the first step in protecting them. Automated tools can discover all IoT devices on the network and diagnose their security posture.
• Configuration management– Changing default passwords, updating firmware, and applying security patches can ensure that all devices are properly configured and secured to the latest manufacturer standards.
• Continuous monitoring and response– AI-powered security solutions can provide the necessary scale and responsiveness to manage large IoT deployments, implementing continuous monitoring to detect and respond to threats in real time.
• User Education and Policies– AI can power engagement-focused dashboards that notify and educate users about what’s happening on their network, what risks it poses, and provide recommendations on next steps.

The stakes are high, and the time to act is now. For many of us, the thought of being attacked seems distant due to a lack of grandiosity or overconfidence.

We continually see in attacks how clever attackers are able to breach networks through poorly secured IoT devices, potentially using them to attack critical infrastructure, public services, and even their own network.