KISTI Launches Development of "Hacking-Zero" Technology for AI-Driven Vulnerability Discovery and Attack Analysis

- Invests 47.2 billion KRW (approx. $35 million) through 2030, with 12 industry, academic, and research entities including ETRI and KAIST participating

- Establishes an AI-driven autonomous cybersecurity system integrating prevention, verification, and analysis of cyberattacks

- Protects national research infrastructure and bolsters South Korea's  competitiveness in AI cybersecurity

□ The Korea Institute of Science and Technology Information (President Sik Lee, hereafter KISTI) announced that it has launched the "Development of a Zero-Hacking System based on AI White-Hat hackers" to proactively defend key national research infrastructure and digital services against sophisticated cyber attacks.

□ The core objective of this initiative is to develop autonomous cybersecurity technology through which artificial intelligence (AI)  identifies and verifies vulnerabilities and attack vectors in information systems, while automatically analyzing attacker behavior and breach processes when security incidents occur. 

□ Cyberattacks have recently evolved at a rapid pace, becoming increasingly sophisticated by exploiting zero-day vulnerabilities and leveraging AI to automate attack processes. If supercomputers, research data, research networks, and various digital services operated by national research institutions fall under attack, it could lead to the disruption of research activities and the leakage of core R&D outcomes and critical national information.

□ However, conventional cybersecurity systems rely heavily on detecting known threats or responding after an incident has already occurred. Furthermore, testing for vulnerabilities within actual operational environments presents significant challenges. Since experts must manually analyze security flaws and attack paths, existing systems face clear limitations in proactively verifying and rapidly responding to newly emerging threats.

□ To overcome these challenges, KISTI, together with 12 industry, academic, and research organizations including the Electronics and Telecommunications Research Institute (ETRI) and the Korea Advanced Institute of Science and Technology (KAIST), will develop a next-generation Hacking-Zero system that integrates AI Cyber Twin, AI White-Hat Hacker, and AI Profiler technologies. The project will run for five years from 2026 to 2030, with a total investment of 47.2 billion KRW.

□ The AI Cyber Twin technology automatically analyzes the architecture, assets, and connectivity of actual networks and information systems to replicate them in a virtual environment. This technology establishes a hybrid living lab where a wide variety of cyberattacks can be safely simulated without directly attacking operational systems or causing service downtime.

□ Within this virtual environment, the AI White-Hat Hacker analyzes systems from the perspective of an actual attacker. By independently exploring vulnerabilities and potential breach paths while repeatedly verifying diverse attack scenarios, the AI identifies latent security risks before real-world attacks occur.

□ The AI Profiler is responsible for post-incident analysis and response. By automatically collecting and correlating logs and digital evidence scattered across multiple information systems, it rapidly identifies attacker behavior, intrusion routes, and attack techniques used, subsequently recommending necessary countermeasures.

□ When these three technologies are integrated, the entire cybersecurity pipeline—from pre-attack prevention to attack verification and post-incident analysis and response—can be executed within a unified system. This enables a paradigm shift from traditional expert-dependent, manual analysis and reactive measures toward an autonomous cybersecurity framework where AI proactively inspects risks, verifies attacks, and diagnoses incident root causes.

□ Dr. Jungsuk Song, Director of the Div. of Network Future Technology Research at KISTI, stated, "This project represents research that utilizes AI to connect the entire cybersecurity lifecycle, ranging from vulnerability discovery to attack verification and incident analysis. We will secure core foundational technologies applicable to national research environments and ensure they lead to enhanced competitiveness for the domestic security industry and research ecosystem."

□ KISTI President Sik Lee stated, "AI is a foundational technology for proactively responding to increasingly intelligent cyber threats. KISTI will actively support industry-academia-research collaboration and R&D so that we can securely protect national research infrastructure and lead the global race in AI cybersecurity technology."