The cyber landscape isn’t getting any easier for businesses to navigate. Attack vectors are growing on all sides, organizations large and small are vulnerable, and malicious actors get more creative and sophisticated every day. But businesses can proactively address security issues by staying on top of evolving cyber threats and shoring up their perimeters accordingly.
Microsoft’s Digital Defense Report from 2021 has shown that the state of cybercrime has many moving parts, but three are stand-out trends that all businesses should be aware of. Malware, malicious domains, and adversarial machine learning are these persistent, innovative threats that are important to stay on top of. Read on to learn more about how they are evolving and what you can do about them.
Malware
You’re likely familiar with the general concept of malware, but it’s a moving target for security professionals because of its tendency to generate new types and evolve. Microsoft’s research discovered that—despite a broad range of outcomes such as data loss and espionage—malware pieces generally rely “on similar strategies for establishing themselves in a network.” Malware types tend to rename payloads to mimic system processes, and cybercriminals tend to use malware to collect data such as credentials from browser caches.
Fileless malware has arrived on the scene, which is malware that “derives most of its components from systems processes or legitimate tools already on a device” according to Microsoft. This makes it more difficult to remove and detect.
Malicious Domains
Malicious domains are any domains used to pursue cybercrime. These can include legitimate websites that have been compromised, or they can be wholly fraudulent infrastructure. The exploding number of domains available on the internet over the last few years has made tracking fraudulent or compromised ones difficult. (Microsoft reported that 30,720 potential new domains are generated in three to four days.)
These domains are typically used as destinations for malware victims, and when victims arrive, they are both initiated into communication with the malicious actor or technology and their location is revealed. Malicious domains can also hide cybercriminals’ identities and locations.
Adversarial Machine Learning
Malicious actors can attack machine learning (ML)-driven models that are originally built to understand, protect, and control data for trusted solutions. When they do, they can open up a new class of vulnerabilities known as adversarial ML. These vulnerabilities can then be exploited to manipulate systems to achieve nefarious goals instead of trusted ones.
Microsoft has incorporated ML-specific security practices including threat detection and mitigation development work and automation into its Security Development Lifecycle to protect its technology against these specific attacks. The company has also discovered that new data is focusing on less noticeable attacks in ML models—finding that tactics like data poisoning actually change the data from which the ML model is built. This changes the very structure on which an ML model is based and can go under the radar until an ML model has a failure.
Protect Your Business in 2023
If you’re concerned about these trends in cybercrime, you are right to be. But businesses don’t have to shrug their shoulders and assume they’ll get hacked. At Point Alliance, we’ve been helping organizations use the right technology to secure their businesses for years, and we’ve kept up with what the hackers are doing. Microsoft has an industry-leading suite of security solutions that we’ve seen demonstrate its power time and time again to keep businesses proactively protected against changing threats. Contact the experts at Point Alliance today to talk about your needs. We’d love to learn how we can help you safeguard your organization.
