A flaw dubbed Java4Shell in OpenJDK, the open-source Java runtime, lets attackers run malicious code without authentication, prompting the presidential office to convene an emergency civilian-government-military meeting on Nov. 17. About 71% of Korea's public information systems are built on Java.
According to the Seoul Shinmun, the vulnerability that researchers named Java4Shell allows attackers to execute malicious code remotely without any account or user authentication. It stems from a memory corruption bug in the Base64 decoder of OpenJDK, the open-source Java development and runtime environment. Base64 is an encoding scheme that converts data into text for network transfer and back again.
광고 문의 · 300×250U.S. security firm Calif, which analyzed the flaw, rated it 9.2 out of 10 under CVSS 4.0, the highest risk category. Oracle released patches on Nov. 17, U.S. time, fixing 11 vulnerabilities in Java SE. The disclosure comes a month after the PHP2Shell flaw, marking another serious defect found at the core of a widely used programming language.
What stands out is how little time and money an attack took to prepare. In Calif's test, one researcher used AI coding tools to analyze Java's core code and found the flaw in about an hour, then built a working exploit in about two hours. The AI token costs came to less than $200, or about 270,000 won. Calif said the most dangerous attacker is a human-AI combination it calls a 'centaur hacker.'
Korea's public sector exposure is broad. Ministry of the Interior and Safety data as of the end of 2025 show that 11,132 of 15,618 public information systems, or 71.3%, were developed in Java. The share was 89.1% for central government agencies and 73.2% for public institutions.
The government has moved to respond. The Ministry of Science and ICT and the Korea Internet and Security Agency urged companies and institutions to run emergency checks and apply security updates, and the National Intelligence Service sent a security advisory on the flaw to public agencies. The presidential office held an emergency meeting with civilian, government and military officials on Nov. 17 to discuss countermeasures. As AI sharply lowers the barrier to finding flaws and writing exploits, patching speed and routine inspection are expected to become decisive for public-sector security.