Microsoft has launched Project Ire, an autonomous AI agent capable of analyzing and classifying software without any human intervention.
Project Ire fully reverse engineers suspicious binaries, reconstructing control flow graphs, running iterative function analysis, and assembling an auditable chain of evidence to determine whether a file is malicious or safe.
In early testing, Project Ire accurately flagged malicious files 98% of the time (with only a 2% false positive rate) when analysing known Windows drivers.
In more challenging tests involving nearly 4,000 previously unclassified files, the tool maintained high precision, correctly identifying about 9 out of 10 malicious files and delivering a low error rate.
With plans to integrate the system into Microsoft Defender as a binary analysis capability, Ire marks a major leap toward scalable, AI-powered cybersecurity automation.
You may also want to check out some of our other recent updates.
Subscribe to Vavoza Insider to access the latest business and marketing insights, news, and trends daily! 🗞️
