The AI Revolution: Reshaping Warfare and Cyber Threats
The rapid advancement of artificial intelligence (AI) is ushering in a new era, drawing parallels to the transformative impact of nuclear fission in the 1940s. AI is poised to fundamentally alter the landscape of conflict, shifting the delicate balance between offensive and defensive capabilities. Its integration into various sectors, from optimising data analysis and logistics to critical decision-making processes, is undeniable. However, its most visible and devastating application to date has been on the modern battlefield, starkly demonstrated in the ongoing conflict between Russia and Ukraine.
AI on the Battlefield: Drones and the Future of Combat
Unmanned Aerial Vehicles (UAVs) have become the vanguard of AI’s military application. These range from nimble quadcopters acting as kamikaze munitions to sophisticated systems capable of launching multiple missiles and returning for replenishment, mirroring the operational flexibility of piloted aircraft. The once-futuristic vision of battles fought by robotically enhanced soldiers and their drone counterparts, operating individually or in coordinated swarms, now appears not just plausible, but probable. AI is undeniably set to play an increasingly pivotal role in future conflicts, heralding a significant revolution in military affairs.
The Escalating Influence of AI in Cyber Warfare
While AI’s impact on physical warfare is significant, its influence on cyber warfare promises to be even more profound and evolve at an accelerated pace. The recent public explosion of awareness surrounding AI is largely attributed to the emergence of powerful generative AI (GenAI). This surge is fuelled by breakthroughs in processing power and sophisticated deep learning models, making GenAI increasingly potent and accessible.
Beyond GenAI, we are witnessing continuous progress in discriminative and predictive AI models. These are fundamental to both AI-driven automation and the cybersecurity domain. Predictive AI has been a cornerstone of cyber threat research and response for over a decade. Organisations like FortiGuard Labs, Fortinet’s dedicated threat research arm, leverage billions of security data points collected daily to derive actionable threat intelligence. While discriminative AI remains crucial for cyber defence, the GenAI revolution acts as a rising tide, elevating capabilities for both legitimate users and malicious actors alike.
The Double-Edged Sword of Generative AI in Cybersecurity
Unlike discriminative AI, which categorises new data, GenAI possesses the remarkable ability to synthesise entirely novel content. This includes text, images, and, critically in the context of cyber threats, computer code. Consequently, GenAI is poised to democratise software development, potentially expanding the pool of exploitable vulnerabilities beyond the current less than five percent of the total. This democratisation will also broaden the variety and diversity of cyberattacks.
Furthermore, GenAI can generate code at a speed far exceeding human programmers. This drastically shrinks the window of opportunity between the public disclosure of a new vulnerability and the emergence of malware designed to exploit it. Historically, organisations might have had days or even weeks to apply vendor-provided patches or workarounds for newly discovered vulnerabilities. However, with GenAI, we may soon witness exploits appearing in the wild within hours or even minutes. Attackers can also leverage GenAI to rapidly analyse new patches and security response tactics, subsequently developing novel variants to circumvent or neutralise defensive measures.
Amplifying Social Engineering and Reconnaissance
Beyond its role in crafting exploits, GenAI is being employed to bolster the credibility of phishing and other social engineering attacks. It enables the creation of increasingly convincing text, voice, and even deep-fake video interactions with unsuspecting victims. The reconnaissance phase of cybercrime, which involves identifying lucrative targets and mapping their networks, is also being transformed by AI. This leads to enhanced breadth, speed, and accuracy in pre-attack examination and assessment of potential targets.
Intelligent, Automated Cybersecurity: Is It Sufficient?
While AI empowers malicious actors to discover vulnerabilities, it also serves as a powerful tool for identifying bugs and mitigating potential security flaws in legitimate software. This is particularly vital given the increasing reliance on open-source and third-party software modules within enterprise supply chains.
Consequently, a growing number of cybersecurity solutions, including those offered by Fortinet, already incorporate AI to assist in identifying vulnerabilities and prioritising their remediation. As the volume of security data continues to expand with the proliferation of digitally connected devices and applications, already stretched security teams facing a global skills shortage can leverage AI-powered automation. This allows them to interpret the deluge of raw data from across their networks and execute the most appropriate responses to anomalous or malicious activities.
Traditional, siloed cybersecurity solutions, often designed to protect against specific threat categories, frequently struggle to counter today’s sophisticated, multi-vector attacks. In contrast, AI-driven solutions that correlate indicators of compromise from multiple data sources and integrate defensive actions across disparate security systems are far better equipped to handle AI-powered assaults.
The Persistent Patching Problem
Despite the clear and substantial advantages of AI-enhanced cybersecurity, a more insidious challenge persists: the fundamental issue of patch management. Even when patches for known software vulnerability exploits are made available, they must be applied to be effective. The “N-day vulnerability” metric, representing the number of days (N) since a vulnerability was disclosed and a patch was released, highlights a critical weakness. Many users fail to apply patches in a timely manner, or at all. Attackers do not always need to develop novel “Zero Day” exploits when lucrative targets neglect to address known vulnerabilities with existing patches. While AI can significantly aid enterprises in vulnerability management, including identification and even automated patching, a concerning number of organisations have yet to fully adopt these capabilities.
Charting the Path Forward
The AI genie is undeniably out of the bottle, destined to reshape our world permanently. To harness its power for beneficial outcomes, optimising autonomy and automation on the digital battlefield with robust, adaptive cybersecurity will be paramount. However, effectively countering the scale and complexity of the current AI-powered threat landscape necessitates a collaborative effort involving industry, governments, and other public bodies.
Companies like Fortinet are at the forefront of innovation, deploying AI-enhanced solutions commercially. This allows governments to concentrate on the technically challenging and rapidly evolving national security implications of AI-powered automation while leveraging the robust capabilities of commercial cybersecurity providers. The key lies in selecting partners with a proven track record of AI innovation, encompassing both generative and predictive technologies, and a steadfast commitment to “Secure by Design” principles – the fundamental belief that strong security must be an integral part of the entire lifecycle of IT products and services, from inception to deployment.
