AI helps uncover a decade-old security flaw in popular programming language Java
LOS ANGELES — A cybersecurity start-up revealed on Tuesday that it used artificial intelligence to uncover a critical flaw in Java, a popular programming language that is used by government agencies and major companies.
It was the latest evidence of a reality facing companies and the government, as AI is revolutionizing the field of discovery in computer vulnerabilities, identifying flaws in code so rapidly that cybersecurity professionals are struggling to keep up.
Using leading AI models, the start-up Calif has now unearthed a string of serious vulnerabilities in recent months.
Java is widely used by banks, hospitals, the aviation industry and technology giants, as well as government agencies, including those with national security missions. It is the world's third-largest programming language used by 90% of the world's Fortune 500 companies like Apple and Amazon.
The discovery comes after a string of serious technology issues, including a massive Microsoft outage and a network failure that grounded all US flights. US Secretary of State Marco Rubio said there is "zero evidence" the incidents are related.
More recently, the US National Security Agency said its internal communications went offline on Tuesday due to a "scheduled routine maintenance".
Apple also reported multiple services suffered full outages due to an internal update.
The NSA did not specify whether its outage was linked to the identified Java bug.
“Java is the beating heart of the modern economy,” said Bruce Dang, a senior researcher at Calif. “Practically speaking, all major companies” will be affected by the bug, he added.
Dang and his colleagues have worked in recent days to coordinate with all government agencies in the intelligence-sharing alliance known as Five Eyes — which consists of the United States, Canada, Australia, New Zealand and Britain — and major vendors in the private sector, including Google and Microsoft.
With the help of AI, Calif discovered a serious security vulnerability called Java4Shell, which allowed an adversary to remotely execute unwanted commands to Java's memory system.
That adversary was OpenAI's recently released model GPT-6 Cyber, which is not yet available to the public. But there are fears hackers could now identify the bug if the flaw isn't fixed.
Oracle, the owner of Java, has not acknowledged the bug, but released a "critical patch update" on Tuesday, which it said contained security vulnerability fixes. At this stage, it is unclear if any platforms or companies have been affected by the latest incident.
But the Java flaw remains a huge concern for cyber-security experts, who say overloading a system's memory is one of the most common ways for hackers to exploit or crash a system.
One of the main functions of Java is that it is supposed to manage and protect memory for you, Professor Toby Walsh, computer scientist and one of the world's leading researchers in AI, told the BBC.
"This is why this is such an unlikely and serious flaw," Walsh added. Java is meant to make the IT world more secure by managing your memory online, he said.
Anytime you transfer information on the internet, receive a file from a user or need to log into an IT system with a password, the platform you are using is likely relying on a Java program.
First released in 1995, the programming language and computing system is now the "beating heart of the modern economy", Calif said in its investigation.
Nearly every critical sector, including banking, national security systems, aviation and telecommunications, relies on Java.
Earlier in November, OpenAI President Greg Brockman announced his company was pivoting 25% of its staff to focus on cyber-security demands.
The announcement followed a critical cyber-security flaw identified in the PHP programming language, identified by Calif.
PHP is the code that websites like Wikipedia use to collect information and present it in a way you can read on the web.
Surrey Centre for Cyber Security Prof Alan Woodward said the PHP flaw affected some 30 million websites, leaving them vulnerable to malicious attacks.
At the time, website owners were urged to check if their sites were affected and, if so, they were told to install updates to patch the issue or consider going offline.
Earlier this week, thousands of customers at major US banks, including JP Morgan Chase, Wells Fargo and Bank of America, reported trouble accessing services.
The latest incidents have exposed how nearly every critical sector — including national security systems, aviation and telecommunications — relies on legacy infrastructure online.
The PHP flaws went undetected for at least a decade, which Walsh said is similar to what has happened with Java.
"It's been hiding in plain sight for 12 years", he said.
"AI is now exposing more flaws much more quickly," Walsh said, including ones that may have been so obscure and difficult that humans could not have discovered them alone.