A US-based security firm that previously helped uncover flaws in PHP has turned artificial intelligence on another cornerstone of the digital economy: Java. Calif announced on Tuesday that its AI-driven analysis found a critical vulnerability in OpenJDK, the most widely used runtime for Java, the world's third most popular programming language.
The bug, dubbed Java4Shell, is a memory corruption issue in OpenJDK's Base64 decoder. Base64 encoding is ubiquitous: it underpins authentication protocols, file uploads, and network message delivery by representing binary data as text. Calif's researchers found that an optimization in the decoder, which uses x86-64 CPU instructions to process multiple bytes simultaneously, can be exploited. Attackers can send specially crafted input that overwrites memory, tricking Java into granting administrator-level access without any password or login.
With that level of control, a remote attacker could execute arbitrary code, install malware, and operate with the same privileges as a company's own IT engineers. The potential impact is vast: Java powers banking and financial systems, cloud and big data infrastructure, aviation, telecommunications, healthcare, and countless connected devices. Around 90% of Fortune 500 companies rely on it, and it remains the backbone of enterprise IT even as newer languages like Python, Go, and Kotlin gain ground in specific niches.
Coordinated response across Europe
Before going public, Calif coordinated with the Five Eyes intelligence alliance—the US, Canada, the UK, Australia, and New Zealand—and major cloud providers to ensure patches were ready before criminals could exploit the flaw. In Europe, national cybersecurity authorities are now mobilising emergency response teams to track the rollout of fixes from Oracle, which leads Java's global maintenance, and from open-source maintainers.
The discovery underscores a growing trend: AI is becoming a powerful tool in both offensive and defensive cybersecurity. Calif's use of machine learning to identify this vulnerability mirrors its earlier work on PHP, which led to patches in October. For European businesses and public institutions, the message is clear: even mature, widely deployed technologies can harbour hidden weaknesses, and the window between discovery and exploitation is shrinking.
European organisations should prioritise applying the patches as soon as they are available. The European Union Agency for Cybersecurity (ENISA) has issued guidance urging member states to assess exposure and update systems promptly. The vulnerability is particularly dangerous because it requires no authentication, making it an attractive target for ransomware gangs and state-sponsored actors alike.
While the immediate focus is on patching, the incident also raises broader questions about the security of open-source software, which underpins much of Europe's digital infrastructure. The European Commission has been pushing for a Cyber Resilience Act that would impose stricter security requirements on digital products, and this episode is likely to reinforce those efforts.
For now, the priority is practical: system administrators across the continent should verify that their Java deployments are updated, monitor vendor advisories, and consider additional network-level protections. The coordinated disclosure model, involving intelligence partners and cloud providers, has given defenders a head start, but the race is far from over.


