SAN DIEGO (NNS) -- Naval Information Warfare Systems Command (NAVWAR) announced the winners of its second annual prize challenge in the Artificial Intelligence Applications to Autonomous Cybersecurity Challenge (AI ATAC) series Dec. 7.
The winner of the challenge and the $500,000 prize was FireEye. Its FireEye Network Security with SmartVision had the best performance based on criteria focused on artificial intelligence (AI) and machine learning (ML) technologies that detect adversarial campaigns by monitoring network-observable behaviors or by analysis of data collected across an enterprise.
Managed by Program Executive Office (PEO) for Command, Control, Communications, Computers and Intelligence (C4I) and Space System’s Cyber Security Program Office (PMW 130), AI ATAC explored adversarial campaign detection using state-of-the-art AI and ML at the network layer. The challenge called for white papers describing network-based security technologies and the corresponding tool for evaluation.
“We are constantly looking for innovative ways to reduce barriers to delivering warfighting capability,” said Rear Adm. Kurt Rothenhaus, program executive officer of PEO C4I and Space Systems. “Prize challenges allows us to bring new folks to the table to solve tough technical challenges with a fresh perspective.”
The program office partnered with Oak Ridge National Laboratory (ORNL), a Department of Energy laboratory in Tennessee, which provided facilities and expertise for evaluating the technical solutions.
Evaluation, using ORNL’s Cybersecurity Operations Research Range, focused on how much of an adversarial campaign, defined as a sequence of events toward an exploitative goal, the challenge participant’s technologies could accurately uncover.
Scoring was based on the estimated cost to an enterprise while using this technology for a given period. The score integrated how much of each campaign was detected, the time to detect, the number of true and false alerts into a simulation of attack costs, labor costs and resource costs. The winning submission was the one with the lowest computed total cost.
“I continue to be amazed at both the high quality and quantity of entries into this challenge series,” said John T. Armantrout, PMW 130 program manager. “Our industry partners are coming to the table with some breakthrough capabilities, and we are gaining great insight on the latest tools available. Congratulations to FireEye for taking home the prize in this challenge – their entry hit it out of the park.
"I’d also like to recognize the fantastic work by my colleague Mike Karlbom as the lead for these challenges; it is no small task to complete one of these challenges."
This was the second AI ATAC prize challenge conducted by NAVWAR enterprise. The first challenge launched in July 2019 and announced the winners in March 2020. FireEye also came out on top for that challenge, which explored the capability for endpoint security products to incorporate AI and ML models to detect and defeat indicators of compromise from various advanced malware strains.
While the winner was the same for both challenges, sponsoring AI ATAC allowed the command to gather important insight and participation from various other nontraditional defense vendors, teams and individuals.
The next prize challenge in the AI ATAC series, to launch later this week, will focus on security orchestration and automated response (SOAR) tools that use AI and ML to automate functions of the Security Operations Center, with a $750,000 prize for the winner. For more information, continue to check www.challenge.gov.