Carnegie Mellon Win Attests to Benefits of Automated Security in DARPA Competition

A start-up that formed at Carnegie Mellon University has won $2 million in the first-ever all-machine hacking tournament. The Las Vegas event was hosted by DARPA, the Department of Defense's Defense Advanced Research Projects Agency, which focuses on national security. The importance of the DARPA Cyber Grand Challenge was that over 12 hours seven teams applied cyber reasoning systems to protect networks and keep applications running autonomously.

Those seven teams were winnowed down from more than a hundred originally playing in earlier qualifying rounds, many of which included the top security researchers and hackers in the world. They're accustomed to playing in highly competitive "capture the flag" tournaments, which test their abilities to evaluate software, seek out vulnerabilities, generate security patches and apply them to computers on the network. But this time the players were machines designed by those kinds of experts, enabling the event to reach its overall goal: to validate the concept of automated cyber defense, bridging the gap between top-notch security software and cutting-edge program analysis research.

"Mayhem," the system built by ForAllSecure, gained the most points as it protected hosts, scanned the network for vulnerabilities and maintained working software.

The automated systems did so well, according to one organizer, they even ended up identifying, locating and protecting against an unknown vulnerability that wasn't part of the game. "In a five-minute window," marveled the organizer in a "highlights" video, "a totally previously never seen before challenge binary was researched, evaluated and found vulnerable and patched by a completely autonomous system."

"Our vision is to check the world's software for exploitable bugs so they can be fixed before attackers use them to hack computers," added David Brumley, CEO of the company behind the winning system as well as director of Carnegie Mellon's CyLab Security and Privacy Institute, and a professor of electrical and computer engineering. "We believe our technology can make the world's computers safe and secure."

ForAllSecure was co-founded in 2012 by Brumley and two Carnegie Mellon graduate students. The startup currently has eight employees and is based in Pittsburgh.

Other teams with institutional representation included Shellphish, which grew out of a security hacking group at the University of California, Santa Barbara; TECHx, a joint University of Virginia and GrammaTech initiative; and CodeJitsu, which brought together researchers from the University of California, Berkeley, Syracuse University and Cyberhaven.

About the Author

Dian Schaffhauser is a former senior contributing editor for 1105 Media's education publications THE Journal, Campus Technology and Spaces4Learning.

Featured

  • abstract glowing circuit patterns

    Microsoft Reduces Copilot Integrations in Windows 11

    Microsoft is dialing back its aggressive Copilot push in Windows 11, promising a sweeping quality overhaul that puts performance and reliability ahead of AI feature expansion .

  • silhouette of business person facing wall of data

    Why AI Strategy Belongs in the President's Office

    Institutions that are succeeding with AI share one thing in common, and it is not a better committee, a larger budget, or a more sophisticated technology stack. It is a president who never handed off the steering wheel.

  • Profile silhouette of a person thoughtfully touching their chin, overlaid with transparent data visualizations and digital interface elements suggesting artificial intelligence and analytics.

    The Institutional Knowledge Shift Is Reshaping Higher Ed IT

    Higher education IT leaders are navigating a quiet but consequential transition: Experienced team members are retiring or leaving for private-sector roles, and the teams replacing them are smaller, newer, and often stretched thin. The result is a structural shift in how technology decisions are made, executed, and sustained.

  • large cloud icon on the right in an abstract world above a polygon with a dark blue background

    Cloud Security Alliance Expands Focus on Governance and Assurance for Agentic AI Systems

    The Cloud Security Alliance (CSA) recently announced a series of CSAI Foundation milestones aimed at securing what it calls the agentic control plane, including a new catastrophic risk initiative, CVE Numbering Authority authorization, and the acquisition of two agentic AI specifications.