TECHX / Xandra
A GrammaTech and University of Virginia Technology
Leading software analysis experts from GrammaTech and UVA came together to compete in DARPA's Cyber Grand Challenge, in which machines played an automated game of capture-the-flag in the name of cyber security research and development.
DARPA challenged the global innovation community with a $2M prize to build a computer that could hack and patch unknown software with no one at the keyboard. On August 4, 2016, DEF CON 24 hosted the world's first all-machine hacking tournament. Seven systems battled against each other using advanced autonomous technologies including: reverse engineering unknown binary software, authoring new IDS signatures, probing the security of opponent software, and re-mixing defended services with machine-generated patches and defenses. GrammaTech's team TECHx – with their robot Xandra – earned 2nd prize, securing a $1M prize.
- Hackers Don’t Have to Be Human Anymore. This Bot Battle Proves It
August 5, 2016 | Wired
- Can machines keep us safe from cyber-attack?
August 2, 2016 | BBC News
- A bunch of computers will try to hack each other in Vegas for a $2 million prize
July 13, 2016 | Tech Insider
- DARPA's Hacking Contest Will Pit Machines Against Each Other
July 5, 2016 | Live Science
- In The Future, We'll Leave Software Bug Hunting to the Machines
June 16, 2016 | Motherboard
- These engineers are developing artificially intelligent hackers
March 3, 2016 | The Guardian
The Story of TECHX and Xandra
A collaboration between GrammaTech and UVA in 2010 led to the development of an automatic software-hardening technology called PEASOUP (Preventing Exploits of Software of Unknown Provenance). Building on this technology, the two groups of software analysis experts came together to compete in DARPA's Cyber Grand Challenge, creating the TECHx system called Xandra.
Xandra competed in CGC's final event on August 4, 2016, and placed 2nd in the competition.
Advancing Research in Autonomic Computing at GrammaTech
Today's software applications have the potential to act erratically after being hacked. To solve this issue, autonomic 'intelligent' software applications operate with self-awareness, monitoring for deviations from expected behavior, assessing mission health, and reacting to preserve the system's functionality – raising alerts, initiating recovery processes, or shutting the system down in order to protect critical data.
Autonomic computing and advanced system reasoning are crucial for IoT and M2M objectives. GrammaTech's research in autonomic computing brings many technologies together to create fully-functioning systems that can detect when they are under attack and protect themselves against exposing critical data or crashing altogether. The PEASOUP and TECHX technologies are great examples of this innovation.