Calif used OpenAI Codex to discover a critical OpenJDK vulnerability.
Original headlineCalif 借助 OpenAI Codex 发现 OpenJDK 严重漏洞
Not investment advice. Muhabbit scores how much a story may matter, not which way to trade.
Security research firm Calif disclosed that it discovered a critical OpenJDK vulnerability, dubbed Java4Shell (CVE-2026-114938), using OpenAI's Codex, with the flaw residing in the Base64 decoder and allowing unauthenticated remote attackers to read data and execute code via memory corruption under certain conditions. Researchers found the bug in roughly one hour and built a working exploit in two more, at a combined token cost of under $200. Exploitation also requires x86-64 processors supporting a specific instruction set, while ARM64 systems are unaffected. OpenJDK has issued a security advisory and Oracle released a patch on November 17, with Java users urged to update promptly. The disclosure highlights how AI coding tools can surface serious flaws quickly and cheaply, and the focus now shifts to patch adoption across Java-dependent infrastructure.
Muhabbit's AI wrote this summary in its own words from TechFlow's report and translated it from Chinese. It can get details wrong, so check the source.
Named in this story
OpenJDK, Calif, OpenAI Codex, Java4Shell, Oracle (ORCL)
A linked name opens that asset's sentiment page.