Artificial intelligence continues to be one of the hot technology topics of the moment for good reason. While there are real-world practical applications already in use, technologists are hyped on how AI (Artificial Intelligence) can augment cybersecurity, especially for IoT devices.
With approximately 7 billion IoT devices deployed today, each one presents an opportunity for hackers to spread malicious software. Also, consider the petabytes of collected data that enable business-critical applications for energy, manufacturing and oil and gas companies, for example.
The loss of data could be crippling to a business, or worse yet, force a business to close altogether. Or consider the fact that breaches are very costly. According to a Capgemini study, “20% (of businesses) report losses of more than 50 million” and “one in five executives said that their organization experienced a cybersecurity breach leading to unauthorized access in 2018.”
The Stuxnet Worm
Consider the Stuxnet worm, which was designed to hunt down nuclear machinery at Iran’s nuclear plant, which has IoT devices connected to centrifuges. After the worm seized control of the centrifuges, it made them spin faster and faster over time, which caused enough strain that the machines disintegrated.
About 20 percent of Iran’s centrifuges were decommissioned as a result of the event. It also likely cost Iran hundreds of thousands, if not millions of dollars and it was all due to a USB-originating worm that was able to infiltrate IoT devices.
In fact, “over a third (35%) of firms said that critical operations (such as website/apps or factory/power grid) were impacted by cyber security breaches” and that “forty-two percent of executives report that cyber security incidents through IoT devices have increased, with the average increase being 16,” according to the same Capgemini study.
It’s events like these that have device managers wondering how secure their devices really are.
Or, device managers often hope for built-in cyber security and leave it at that. But when AI is applied to cyber security, it enables a set of capabilities that allows organizations to predict, detect and respond to cyber threats in real-time versus after the fact.
How AI is Used to Boost Cyber Security
AI-based algorithms can be deployed directly on IoT devices, which run pattern recognition to identify ransomware or malware attacks. The AI then can also isolate the threat. Through alerts, an IoT device manager would be notified that there is a potential breach present and they would be able to prevent it from being activated or from spreading to other devices.
With an AI-driven cyber security program, device managers are also able to lower the cost to detect and respond to breaches, respond faster to breaches and spend less time combing through data logs.
Simply put, the threat is now more real than ever and technologists need to get on board with using AI for breach protection.
How to Implement AI in Cyber Security
- Appoint an internal cyber security champion or hire a consulting firm
First and foremost, you’ll need either an AI internal champion or outside counsel to get your project off the ground. Without the expertise you need or a dedicated champion, you’ll undoubtedly struggle to maintain a focus on AI. With IoT deployments being so complex in nature, it’s important to have an individual championing the process within your company.
2. Determine the best use cases
Naturally, you’ll need to determine which areas of your business can benefit the most from AI-driven cyber security. It may not make sense to apply AI to every type of cyber security measure you have in place. Perform a cyber security audit on all your devices to determine any weaknesses you have and determine where it makes the most sense to implement AI.
3. Stay on top of emerging cyber security threats
Just as you are doing everything you can to stop hackers, they too are always evolving. Stay on top of the latest cyber security threats by reading up on the topic. Visit forums to collaborate with other cyber security specialists in order to keep up with the type of threats that others in the space are dealing with. The hackers are constantly evolving, so you should be too.
4. Perform regular audits of IoT hardware
Once you do begin using AI to drive your cyber security program, it’s important to regularly audit software and devices to ensure everything is working properly. While AI can help you detect threats before they happen, it’s still wise to regularly audit your devices and cyber security practices to ensure everything is working as designed.
5. Enable governance
Governance, the use of management controls to make sure that IT functions as designed, is a critical aspect of your cyber security program. Governance will ensure you have strategic buy-in, the AI produces the outcomes expected and that it hasn’t been compromised. Given the complex nature of IoT deployments, governance is a great way to ensure your project fulfills your original business goals.
6. Have a recovery plan in place
If you do have a breach, be sure to have a recovery plan in place. Review the incident to determine how you could improve security measures going forward. Once you have implemented changes to your cyber security program, test again to ensure your practices are producing the intended outcome.
By following these steps, you’ll ensure that your devices and cyber security practices are always up to date. However, something to keep in mind is that hackers also use AI to help deploy attacks, which is why it’s important to stay on top of cyber security. The future of cyber security is AI. Are you ready?