Articles

CSIAC collects and publishes articles related to our technical focus areas on the web to share with the DoD community.

Filter by Technical Focus Areas

By default, only content within your selected technical focus areas is displayed throughout the site. You can update your technical focus areas in your profile or temporarily filter the content here.


Filter by Article Types
graphic red network lines

NSA, FBI, CISA, and Allies Issue Advisory About Russian Military Cyber Actors

FORT MEADE, Md. – The National Security Agency (NSA) joins the Federal Bureau of Investigation (FBI), the Cybersecurity and Infrastructure…

eight-inch wafer that can selectively reflect different colors of light

Lighting the Way for Quantum Innovation

ALBUQUERQUE, N.M. — Sandia National Laboratories and Arizona State University, two research powerhouses, are collaborating to push the boundaries of…

ORNL's Moe Khaleel, left, associate laboratory director for national security sciences, and Maurice Singleton III, chief executive officer of U2opia Technology, celebrate the partnership between Oak Ridge National Laboratory and U2opia Technology. Credit: Carlos Jones/ORNL, U.S. Dept. of Energy

U2opia Signs License to Commercialize Anomaly-Detection Technology for Cybersecurity

U2opia Technology has licensed Situ and Heartbeat, a package of technologies from the Department of Energy’s Oak Ridge National Laboratory…

Air Force Gen. Timothy Haugh, commander of U.S. Cyber Command and director of the National Security Agency, gives the opening remarks at the Cyber Command’s Legal Conference at Joint Base Andrews, Md., Aug. 9, 2024.

AI Security Center Keeps DoD at Cusp of Rapidly Emerging Technology

The director of the National Security Agency said the agency’s new Artificial Intelligence Security Center is paying dividends in the…

U.S. Cyber Command seal

USCYBERCOM Unveils AI Roadmap for Cyber Operations

FORT GEORGE G. MEADE, Md.   –  Mr. Michael Clark, deputy director of plans and policy at U.S. Cyber Command, presented a…

codebreaker challenge graphic

NSA Launches 11th Annual Codebreaker Challenge for 2024

FORT MEADE, Md. – The National Security Agency (NSA) is launching its annual Codebreaker Challenge, offering students from U.S.-based academic…

binary-cad-design-of-part

CRWS-BoK Portal

The U.S. Department of Defense announced the release of Version 4.2 of the Online Cyber Resilient Weapon Systems Body of…

city-skyline-network

CISA Releases Analysis of FY23 Risk and Vulnerability Assessments

The Cybersecurity & Infrastructure Security Agency (CISA) has released an analysis and infographic detailing the findings from the 143 risk…

graphic of a scale

Artificial Intelligence in Cybersecurity: How Substituting and Scaling Impact Investment Returns

With the growing integration of artificial intelligence (AI) in cybersecurity, this article investigates the economic principles of substitution and scale’s elasticity to evaluate their impact on the return on security investment. Recognizing the potential of AI technologies to substitute human labor and traditional cybersecurity mechanisms and the significance of cost ramifications of scaling AI solutions within cybersecurity frameworks, the study theoretically contributes to understanding the financial and operational dynamics of AI in cybersecurity. It provides valuable insights for cybersecurity practitioners in public and private sectors. Through this analysis, ways in which AI technologies can redefine economic outcomes in cybersecurity efforts are highlighted. Strategic recommendations are also offered for practitioners to optimize the economic efficiency and effectiveness of AI in cybersecurity, emphasizing a dynamic, nuanced approach to AI investment and deployment.

graphic of a satellite

Optimization and Analysis for Defense Simulation Models

When performing defense system analysis with simulation models, a great deal of time and effort is expended creating representations of real-world scenarios in U.S. Department of Defense (DoD) simulation tools. However, once these models have been created and validated, analysts rarely retrieve all the knowledge and insights that the models may yield and are limited to simple explorations because they do not have the time and training to perform more complex analyses manually. Additionally, they do not have software integrated with their simulation tools to automate these analyses and retrieve all the knowledge and insights available from their models.

black and white graphic of a soldier in combat gear

Navigating Challenges and Opportunities in the Cyber Domain With Sim2real Techniques

In the digital age, the cyber domain has become an intricate network of systems and interactions that underpin modern society. Sim2Real techniques, originally developed with notable success in domains such as robotics and autonomous driving, have gained recognition for their remarkable ability to bridge the gap between simulated environments and real-world applications. While their primary applications have thrived in these domains, their potential implications and applications within the broader cyber domain remain relatively unexplored. This article examines the emerging intersection of Sim2Real techniques and the cyber realm, exploring their challenges, potential applications, and significance in enhancing our understanding of this complex landscape.

graphic of a microchip with electric pulses coming from it

Low-Power Cybersecurity Attack Detection Using Deep Learning on Neuromorphic Technologies

Neuromorphic computing systems are desirable for several applications because they achieve similar accuracy to graphic processing unit (GPU)-based systems while consuming a fraction of the size, weight, power, and cost (SWaP-C). Because of this, the feasibility of developing a real-time cybersecurity system for high-performance computing (HPC) environments using full precision/GPU and reduced precision/neuromorphic technologies was previously investigated. This work was the first to compare the performance of full precision and neuromorphic computing on the same data and neural network and Intel and BrainChip neuromorphic offerings. Results were promising, with up to 93.7% accuracy in multiclass classification—eight attack types and one benign class.