Skip to content
  • Blog Home
  • Cyber Map
  • About Us – Contact
  • Disclaimer
  • Terms and Rules
  • Privacy Policy
Cyber Web Spider Blog – News

Cyber Web Spider Blog – News

Globe Threat Map provides a real-time, interactive 3D visualization of global cyber threats. Monitor DDoS attacks, malware, and hacking attempts with geo-located arcs on a rotating globe. Stay informed with live logs and archive stats.

  • Home
  • Cyber Map
  • Cyber Security News
  • Security Week News
  • The Hacker News
  • How To?
  • Toggle search form

Microsoft’s Project Ire Autonomously Reverse Engineers Software to Find Malware

Posted on August 5, 2025August 5, 2025 By CWS

Microsoft on Tuesday unveiled Undertaking Ire, a prototype autonomous AI agent that may analyze software program information with a view to decide whether or not they disguise malware.

Based on Microsoft, Undertaking Ire can autonomously reverse engineer and classify software program with none prior context, automating and scaling what could be a complicated course of. 

Undertaking Ire was developed by groups at Microsoft Analysis, Microsoft Defender Analysis, and Microsoft Discovery & Quantum. 

It makes use of decompilers and different instruments to collect knowledge that permits it to find out whether or not a file is benign or malicious, whereas additionally offering a traceable chain of proof.

“The system’s structure permits for reasoning at a number of ranges, from low-level binary evaluation to regulate stream reconstruction and high-level interpretation of code habits,” Microsoft defined.

It added, “Its tool-use API permits the system to replace its understanding of a file utilizing a variety of reverse engineering instruments, together with Microsoft reminiscence evaluation sandboxes primarily based on Undertaking Freta, customized and open-source instruments, documentation search, and a number of decompilers.”

Undertaking Ire’s purpose is to scale back analyst error and fatigue, speed up risk response, and strengthen defenses towards evolving assaults, Microsoft stated.

In checks performed by the tech large on a dataset of Home windows drivers that included each malicious and benign software program, Undertaking Ire accurately recognized 90% of information and solely flagged 2% of the benign information as harmful. Commercial. Scroll to proceed studying.

In a special take a look at focusing on roughly 4,000 information that had been lined up for reverse engineering and evaluation by human consultants, Undertaking Ire accurately flagged 9 out of 10 malicious information as malicious, with a false constructive price of solely 4%. Nonetheless, it was solely in a position to detect roughly 1 / 4 of all precise malware. 

Microsoft admitted that the general efficiency was reasonable, however argued that the testing circumstances have been difficult and the outcomes nonetheless point out “actual potential for future deployment”.

“Primarily based on these early successes, the Undertaking Ire prototype can be leveraged inside Microsoft’s Defender group as Binary Analyzer for risk detection and software program classification,” Microsoft stated.

“Our purpose is to scale the system’s pace and accuracy in order that it could accurately classify information from any supply, even on first encounter. In the end, our imaginative and prescient is to detect novel malware immediately in reminiscence, at scale,” it added.

Associated: Microsoft Provides $5 Million at Zero Day Quest Hacking Contest

Associated: Microsoft Boosts .NET Bounty Program Rewards to $40,000

Associated: Microsoft to Preview New Home windows Endpoint Safety Platform After CrowdStrike Outage

Security Week News Tags:Autonomously, Engineers, Find, Ire, Malware, Microsofts, Project, Reverse, Software

Post navigation

Previous Post: New Streamlit Vulnerability Allows Hackers to Launch Cloud Account Takeover Attacks
Next Post: ClickFix Malware Campaign Exploits CAPTCHAs to Spread Cross-Platform Infections

Related Posts

DDoS Attacks Blocked by Cloudflare in 2025 Already Surpass 2024 Total  Security Week News
Jaguar Land Rover Says Shutdown Will Continue Until at Least Oct 1 After Cyberattack Security Week News
Tidal Cyber Raises $10 Million for CTI and Adversary Behavior Platform Security Week News
Pixnapping Attack Steals Data From Google, Samsung Android Phones Security Week News
Asus DriverHub Vulnerabilities Expose Users to Remote Code Execution Attacks Security Week News
Cyberattack on Beer Giant Asahi Disrupts Production  Security Week News

Categories

  • Cyber Security News
  • How To?
  • Security Week News
  • The Hacker News

Recent Posts

  • Exploited ‘Post SMTP’ Plugin Flaw Exposes WordPress Sites to Takeover 
  • Mysterious ‘SmudgedSerpent’ Hackers Target U.S. Policy Experts Amid Iran–Israel Tensions
  • 239 Malicious Android Apps on Google Play With Downloaded Over 40 Million Times
  • Jupyter Misconfiguration Flaw Allow Attackers to Escalate Privileges as Root User
  • Cybersecurity Professionals Charged for Deploying ALPHV BlackCat Ransomware Against US Companies

Pages

  • About Us – Contact
  • Disclaimer
  • Privacy Policy
  • Terms and Rules

Archives

  • November 2025
  • October 2025
  • September 2025
  • August 2025
  • July 2025
  • June 2025
  • May 2025

Recent Posts

  • Exploited ‘Post SMTP’ Plugin Flaw Exposes WordPress Sites to Takeover 
  • Mysterious ‘SmudgedSerpent’ Hackers Target U.S. Policy Experts Amid Iran–Israel Tensions
  • 239 Malicious Android Apps on Google Play With Downloaded Over 40 Million Times
  • Jupyter Misconfiguration Flaw Allow Attackers to Escalate Privileges as Root User
  • Cybersecurity Professionals Charged for Deploying ALPHV BlackCat Ransomware Against US Companies

Pages

  • About Us – Contact
  • Disclaimer
  • Privacy Policy
  • Terms and Rules

Categories

  • Cyber Security News
  • How To?
  • Security Week News
  • The Hacker News