Anand Oswal, SVP & GM of Network Security, Palo Alto Networks.
Every day, I hear from government and company leaders asking how they can best leverage artificial intelligence (AI) and machine learning (ML) for improved cybersecurity outcomes—and for protection against more sophisticated attacks created by bad actors using AI. What’s common among these leaders are three key challenges:
• Most companies are using, on average, 45 different cybersecurity tools, leading to a host of issues, including gaps in security coverage, misconfigurations and reliance on manual work.
• Add to that the fact that the industry is facing a cybersecurity talent shortage, which means that finding the right skill sets and training and retaining security professionals to support these different tools can be challenging.
• Finally, companies are losing precious insights from the data that’s stuck in little pockets of cybersecurity tools, which means the data isn’t being intelligently leveraged for holistic cybersecurity risk management.
Now I’m going to let you in on the big secret to leveraging AI for cybersecurity technology—which, by the way, also addresses all of these challenges faced by most CISOs. That secret is to seize the benefits of platformization. Platformization creates integration and interoperability between security solutions, leading to improved visibility and control over the security infrastructure. Different stacks from different vendors make cybersecurity unnecessarily complex, whereas platformization creates a view across your ecosystem through a single pane of glass. This allows for unified management and operations, so you can write policy from one place and enforce it everywhere, consistently, through quick cloud-based deployment.
• Contextual Understanding: Combining data from different sources provides a more comprehensive understanding of the causality surrounding security events. For example, correlating network traffic logs with endpoint behavior and cloud activity can help identify potential indicators of compromise (IOCs) and uncover hidden threats that may be missed when analyzing data in isolation.
• Real-Time Attack Prevention: AI models trained on diverse datasets can learn complex patterns and behaviors across multiple domains. This enables them to detect advanced threats, such as sophisticated malware, insider threats or coordinated attacks, by identifying anomalies and abnormal activities that might not be evident within a single data source. This allows for better detection and faster response.
The fact is that good AI comes from good data. When it comes to supercharging cybersecurity with AI, the more breadth and depth of data the model has access to, the better it will perform. That’s why the integration of data from various sources and domains is vital. It’s about stitching together information from different sources to provide a holistic view of the security landscape. As the model learns, it gains value from the network effect—sharing insights across enforcement points.
It’s clear that companies have a sense of urgency when it comes to aligning their security strategy while taking stock of their legacy systems currently in place. Those that aren’t keeping an eye on the horizon and the velocity at which network security vulnerabilities evolve will find themselves lagging. Cybersecurity isn’t static, so we must consistently be working to improve our security posture.
As you reduce the complexity of your operations by streamlining the number of tools and vendors, it makes it easier to manage the environment, train your staff, control costs, quickly identify and respond to risks, and create better security outcomes—which matters the most. And now you’re in on the big secret.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
Read the full article here
