Microsoft has unveiled a new AI agent named Project Ire, which is able to classify malware at a global scale with precision.
Announced during Black Hat USA 2025, Project Ire is capable of fully reverse engineering a software file without any clues about its origin or purpose. It uses decompilers and other tools to review the software’s output to determine whether it is malicious or benign.
The system uses advanced language models and a suite of callable reverse engineering and binary analysis tools to drive investigation and adjudication.
The prototype agent has demonstrated its efficacy across a series of tests, including a real-world scenario of around 4000 “hard-target” files not classified by automated systems.
To date, Project Ire has achieved a precision of 0.98 and a recall of 0.83 using public datasets of Windows drivers, according to a Microsoft blog dated August 5.
“It was the first reverse engineer at Microsoft, human or machine, to author a conviction case – a detection strong enough to justify automatic blocking – for a specific advanced persistent threat (APT) malware sample, which has since been identified and blocked by Microsoft Defender,” the Project Ire researchers wrote.
Project Ire is also able to invoke a validator tool that cross-checks its initial findings. This tool draws on expert statements from malware reverse engineers on the Project Ire team. Drawing on this evidence and its internal model, the system creates a final report and classifies the sample as malicious or benign.
Various Microsoft teams collaborated in the development of the AI agent including those with security expertise and operational knowledge alongside data from global malware telemetry and AI research.
Project Ire Available to Microsoft Customers
Following the successful preliminary tests, the Project Ire prototype will be leveraged inside Microsoft’s Defender organization as a binary analyzer for threat detection and software classification.
“Our goal is to scale the system’s speed and accuracy so that it can correctly classify files from any source, even on first encounter. Ultimately, our vision is to detect novel malware directly in memory, at scale,” the researchers noted.
The agent aims to help address burnout and alert fatigue experienced by security analysts.
Traditional AI malware analysis tools cannot clearly indicate whether a sample is malicious or benign, meaning analysts are forced to investigate each sample incrementally, Microsoft noted.
