Open-source CyberSOCEval transforms malware testing

A new open-source platform called CyberSOCEval is drawing attention in the cybersecurity field for showing how artificial intelligence can enhance malware detection and threat intelligence, GBHackers News reports.

Built by independent researchers, the tool tests AI engines against a curated library of malware families ranging from trojans to ransomware, scoring them on accuracy, false positives, and speed. According to early trials, several open-source AI models matched or outperformed commercial offerings, with the platform providing what researchers describe as a "transparent benchmark" for evaluating performance.

CyberSOCEval goes beyond detection by assessing how effectively AI can generate actionable intelligence, including mapping attack chains, analyzing network traffic, and linking command-and-control infrastructure. In one case, it cut manual effort in mapping attacks by more than 60 percent.

All code, malware samples, and test protocols are freely available on GitHub, allowing researchers, vendors, and academics to contribute. Advocates say this community-driven model ensures benchmarks stay current while pushing the development of stronger AI-driven defenses.