As adoption of the Internet of Things (IoT) and the Industrial Internet of Things (IIoT) ramps up, the cybersecurity threat landscape changes from PCs, tablets, and conventional networks to all sorts of connected devices, including large, durable goods (think airplanes, automobiles, and construction equipment). IIoT connects major equipment such as aircraft, automobiles, and critical pieces of national infrastructure, including turbines used in power generation and transformers and switches in our electricity grid.
Cisco estimates that the number of devices connected to IP networks will be more than three times the global population by 2022, in the range of 30 billion to 50 billion, with a projected annual economic impact of $3.9 trillion to $11.1 trillion worldwide by 2025. Even if growth projections are overly optimistic, cybersecurity must assume a much more critical role. If a piece of durable machinery becomes infected, the consequences pose a serious threat to business performance and production, as well as overall safety — and could even have national security implications. As older, unconnected equipment is replaced with new, connected machinery, fears are beginning to rise about how secure this new equipment is and what can be done to make it more impregnable from cyberattacks.
Here are three things that original equipment manufacturers (OEMs) should focus on to enhance data security, while also fully realizing the promised economic benefits of IoT and IIoT.
1. Adopt Industrywide Standards
In the software world, we design with security in mind and adhere to a set of well-defined and mature industry standards that allow for different parts of the technology stack to interact with each other. However, the industrial manufacturing world lacks overarching standards for IoT. Instead, in certain industries, there is often competition between large players who rely on a small set of proprietary and incompatible technology standards. This lack of common standards makes it difficult to develop end-to-end security. The manufacturing industry needs to follow the software industry’s lead and come together to define standards. Equipment is becoming increasingly complex with millions of lines of code and the introduction of more equipment that’s designed with software in mind. OEMs must work more closely with suppliers and industry organizations to accelerate the development of industry standards for the greater benefit of everyone.
2. Improve Communication and Sophistication
In the manufacturing world, specifically the operations technology (OT) sphere, legacy operational standards such as OPC and Modbus are still in use today but were designed more than 20 years ago using old technologies, including COM. They were not designed for communication over modern IP networks with multiple security layers and, due to a general lack of cybersecurity sophistication, traditional OT networks have most security options disabled to simplify configuration.
By its nature, a large open network of connected devices opens many new attack vector threats, even if individual devices may be secure when used independently. Because the weakest point in the system determines its overall security level, a comprehensive end-to-end approach is required to secure it. The lack of industry standards within the manufacturing space makes it difficult to develop such an approach because hackers concentrate on breaching a specific element within the technology stack.
To combat this, manufacturers must adopt a similar standard to what’s found within the software industry, where communication networks are completely closed and ports are only opened as needed, and comprehensive end-to-end approaches are designed. Traditional industrial component suppliers and OEMs are not well positioned to perform this task, but the industry has to improve its level of sophistication to provide end-to-end protection against all types of attack vector threats.
3. Increased Focus on On-Device Security, Leveraging IoT Strengths
While network-level or cloud infrastructure cybersecurity is very important, manufacturers also must realize that device-level cybersecurity is equally or more important. Most data is still stored in the cloud, yet manufacturers must come to terms with data and information also being stored directly on devices themselves. This means a "cloud only" cybersecurity strategy won’t get the job done.
Endpoint security solutions designed for smartphones and tablets can't be expected to work for IoT sensors and devices because the design of IoT devices makes this unfeasible at a typical enterprise level. Manufacturers need a combination of network-level security and solutions tailored specifically for IoT architectures, including sensor devices. One of the biggest strengths of IoT is the sheer amount of data that devices are expected to generate, and this same strength could be used to apply machine learning models to detect anomalous malicious behavior that could compromise security.
IoT technology is one of the most exciting disruptions to hit the manufacturing space in years. However, security has lagged behind technology adoption, and if we don't r-think and adopt a new comprehensive security model for connected devices, the value that IoT promises to deliver will just remain that — a promise. By implementing the steps above, OEMs and industry can better prepare for cybersecurity challenges and have more peace of mind when adopting IoT technology.
Check out The Edge, Dark Reading's new section for features, threat data, and in-depth perspectives. Today's top story: "The Right to Be Patched: How Sentient Robots Will Change InfoSec Management."