November 18, 2026 USD 45.34EUR 49.86PLN 11.32
Новини Харків — коротко й по суті
LIVE
European Union Prepares Multi-Billion Funding Amid Russian Plans for Millions of Drones Zelenskyy strongly condemns Russian strike on Odesa, says escalation disproves peace claims "Karthage" Case: High Anti-Corruption Court Arrests NABU Tapes Figure Stas Shurins reveals reasons for leaving Ukraine Remote Mining in Kharkiv Region: Explosives Found Hidden in a Handbag Why indoor plant leaf tips dry out: Experts debunk common watering myths Ukrzagaliznytsia Tariff Hikes Threaten Regional Industry and Logistics Ukrainian Forces Strike Bridge Over Siverskyi Donets and Russian Military Infrastructure New Critical Java Vulnerability: Artificial Intelligence Helps Uncover Dangerous Flaw Verkhovna Rada Reports Decrease in Street Conflicts During Mobilization European Union Prepares Multi-Billion Funding Amid Russian Plans for Millions of Drones Zelenskyy strongly condemns Russian strike on Odesa, says escalation disproves peace claims "Karthage" Case: High Anti-Corruption Court Arrests NABU Tapes Figure Stas Shurins reveals reasons for leaving Ukraine Remote Mining in Kharkiv Region: Explosives Found Hidden in a Handbag Why indoor plant leaf tips dry out: Experts debunk common watering myths Ukrzagaliznytsia Tariff Hikes Threaten Regional Industry and Logistics Ukrainian Forces Strike Bridge Over Siverskyi Donets and Russian Military Infrastructure New Critical Java Vulnerability: Artificial Intelligence Helps Uncover Dangerous Flaw Verkhovna Rada Reports Decrease in Street Conflicts During Mobilization

New Critical Java Vulnerability: Artificial Intelligence Helps Uncover Dangerous Flaw

💡 Tech 2 min read 👁 0

Researchers using a specialized AI model have discovered a dangerous vulnerability in the Java programming language affecting thousands of companies and government institutions worldwide.

New Critical Java Vulnerability: Artificial Intelligence Helps Uncover Dangerous Flaw💡Tech

Photo: RBC-Ukraine

Key points

  • Critical vulnerability discovered in Java programming language using AI
  • The attack targets widely used management systems deployed by global organizations
  • Developers have released security patches, but outdated systems remain at risk

Another alarming cybersecurity precedent has emerged as cutting-edge artificial intelligence technologies helped uncover a critical vulnerability in the Java programming language. The issue affects a system widely used by government structures and the vast majority of the world's leading corporations. This discovery once again highlights the rapid evolution of methods for finding system flaws using next-generation algorithms.

Details of the Discovery and Scale of the Threat

Security researcher Khang Luong successfully identified the dangerous bug while thoroughly analyzing the standard base64 decoder, designed to convert binary data into text. It turned out that a specific combination of characters under certain conditions leads to incorrect data processing in computer memory. The Calif team engaged OpenAI's specialized GPT-6 Cyber model to turn this glitch into a full-fledged attack vector and gain system control.

During test runs, the popular Keycloak identity and access management tool—used by prominent global organizations including NASA and CERN—was targeted. Independent journalist verification confirmed that carrying out such an attack takes less than an hour. Although developers at Oracle have already released a corresponding fix and security agencies have urged urgent software updates, the situation remains tense for legacy systems that do not receive regular patches.

"I don't know how much time we have left, maybe three months, maybe three years. Right now, the best hacker is a centaur hacker," noted Calif CEO Tai Duong.

Source RBC-Ukraine — rbc.ua/ukr/news/gpt-6-cyber-znayshla-nebezpechnu-vrazlivist-1795001… ↗

Contact the newsroom

Got a tip, a proposal or a partnership question? Write to us and we will reply shortly.