By Vokreka Senatus, Warrior Cyberwar Analyst
The integration of Artificial Intelligence (AI) into cybersecurity is paramount for enhancing the defense against evolving cyber threats, something which is particularly crucial in military environments, such as armored vehicles amid heavy combat, air-to-ground networking and targeting with fighter jets, and naval environments like the USS Gerald R. Ford. These are all combat circumstances where harsh environmental conditions pose unique challenges to data collection and processing. Both weather obscurants and bandwidth challenges can complicate the growing extent to which AI can safeguard data transiting between platforms and across domains. Should air, land or sea combat conditions impact the environment, the scope and quality of data aggregated by various sensors can be impacted in a way that challenges information analysis and transition.
Environmental and combat variables can also impact bandwidth, something which naturally impacts the volume, speed and efficiency of data collection, organization and transmission. Perhaps fog, snow or sandstorms impact the resolution of EO/IR-collected sensor data, or “electromagnetic jamming” or cluttering compromises bandwidth? Both of these scenarios would benefit from AI-enabled algorithms designed to accommodate fluctuations and changing variables impacting data collection and processing.
Therefore, the services have been working intensely to improve the efficiency of data collection and analysis and ensure fidelity of the data itself amidst changing circumstances.
These efforts have both gained momentum and achieved great success in recent years and are consistently being enhanced. For instance, nearly a decade ago, US Navy cybersecurity experts told Warrior about a special Task Force Cyber Awakening designed to increase computer security and harden networks given the fast-growing extent to which weapons systems and major combat platforms are becoming computer reliant. This is something AI can greatly improve, yet it seems it must be applied and engineered in a way that accommodates or accounts for these kinds of challenges and variations.
Part of Task Force Cyber Awakening’s focus was to improve cyber awareness and sophistication, recognizing the growing reliance on AI-enabled programming in weapon platforms, and it is an effort which continues to evolve into today’s modern environment. Today, systems need engineering to accommodate for changing variables in critical areas such as impact of combat circumstances, bandwidth limitations, sensoring, and center data flow and data transmission. Additionally, AI systems should be robust enough to handle these variations in network traffic and potential disruptions caused by any environmental factors analogous to detection, response, and mitigation of cyber threats; providing a proactive defense against evolving security challenges.
Integrating AI into cybersecurity for military applications requires a comprehensive understanding of both the cybersecurity landscape and the unique challenges posed by harsh environmental conditions in naval operations. Rigorous testing and validation processes are essential to ensure the reliability and effectiveness of AI-based cybersecurity solutions in such demanding scenarios. Theoretically, during operations, if a multitude of electronic systems are all functioning at high-volume or full-capacity, then the risk for electronic warfare or breach is respectfully understood to also be positively reinforced. Therefore, as long as the USS Ford can experience vibrations or electromagnetic shocks during operations, this should be recognized as a risk. AI components must be ruggedized to withstand such any physical stresses.
Analogous to the consideration of physical military environments, exclusively throughout naval operations, AI hardware and systems must also be designed to reliably operate under extreme measures and still sustain capacity to continuously log data analytics in accordance to the systems’ exposure. Without proper mitigation protocols, the possibility of breach increases. Any possible signal interruptions allude to electromagnetic interference; this can register troubleshooting as limited bandwidth. It can result from factors such as network congestion, low available data rates, or intentional attack thereby operating through the potential bandwidth limitation. Limited bandwidth can impact the speed and efficiency of data transfer, leading to delays, reduced performance, or the inability to transmit large amounts of data quickly. Especially in scenarios where communication channels are constrained, automations in connectivity data must be pre scripted, encrypted, and readily on standby.
AI-powered endpoint protection can detect and respond to threats on individual devices connected to the network. Endpoints in military environments may face unique challenges, including exposure to extreme temperatures and potential physical damage. AI can automate responses to certain security incidents, enabling faster reaction times. Those automated responses should be designed to operate effectively in the presence of environmental disturbances, ensuring they don’t inadvertently exacerbate issues. AI-based systems can analyze emails and other communications to detect phishing attempts and malicious content. Robust AI algorithms are needed to adapt to variations in communication patterns and potential disruptions to regular communication channels. AI tools can conduct automated vulnerability assessments to identify weaknesses in the network infrastructure and the assessments should consider referral to potential vulnerabilities introduced by harsh environmental conditions, such as physical damage or corrosion.
Military operations may involve periods of limited connectivity, and AI systems should be able to detect its limitations and operate autonomously within reduced connectivity. Unstable
power sources or sudden power fluctuations can impact AI systems. Implementing robust power management systems and backup solutions are crucial in a general sense and especially during times of limited connectivity to power source(s).
In conclusion, the effectiveness of AI in cybersecurity for military applications lies in its adaptability to recognize changing patterns, resiliency in harsh environmental conditions, and the ability to operate autonomously in challenging scenarios. While performing behavioral analysis to register and identify normal vs. abnormal user activities, and predict possibilities of exploit that may indicate a security breach, the AI system needs to adapt to any changing patterns influenced by operational needs and external conditions in real-time.
By Volreka F. Senatus is a Warrior Cyberwar analyst, student and Assistant Educator at Florida International University
