14 Sources
[1]
AI helps uncover critical flaw in ubiquitous internet software
If you checked your bank balance, booked a flight or ordered groceries online today, there is a good chance your request was handled by a computer program written in Java. For nearly three decades, the Java programming language has been one of the quiet workhorses of the internet. While most
[2]
US cyber security agency warns over vulnerability in Oracle's Java
Companies told to update systems running on programming language widely used by corporate sector, banks and government agencies Oracle's popular Java software is vulnerable to a potentially serious cyber attack, a US cyber security agency has warned, raising fears of further web outages. The
[3]
AI found a hole in Java, and it starts with 'AAAAAAAA...'
The bug could let hackers take over systems at banks, hospitals and most Fortune 500 companies. If the last week is any indication, the coming months and years will be a wild ride for digital infrastructure. After outages at Microsoft, the FAA, banks, Apple and... seemingly everything else, we've
[4]
A.I. Helps Find Major Security Bug in Java, a Software Platform Vital to the Internet
The discovery by security company Calif shows A.I.'s ability to find gaping security holes in popular software. The Java flaw could affect systems used by nearly all Fortune 500 companies. A cybersecurity start-up revealed on Tuesday it had used a new A.I. model to uncover a critical software bug
[5]
Critical flaw found in vital internet platform Java through AI
A cyber-security company has used artificial intelligence (AI) to uncover a security flaw in the hugely popular computer language program Java. The tech start-up Calif - which recently identified a cyber-security flaw in another programming language, PHP - found the critical software bug in Java's
[6]
AI is now finding bugs in the code that runs banks, airlines, and hospitals
Serving tech enthusiasts for over 25 years. TechSpot means tech analysis and advice you can trust. The takeaway: An AI model built for security research has uncovered a series of vulnerabilities in Java, the Oracle programming language used in the systems of airlines, banks, hospitals, and
[7]
'Java is everywhere': AI-found Java4Shell flaw hits Jira and Jenkins
Calif says it found Java4Shell in an hour with about $200 of AI tokens, and the trigger had been public since September. Attackers are already probing, and CISA says to patch now. A critical flaw in Java can let attackers run their own code on widely used business software without logging in.
[8]
Java4Shell: AI found a critical Java flaw, and Oracle patched it
Security firm Calif disclosed a critical flaw in OpenJDK, the most popular Java runtime, that it says lets an attacker with no credentials overwrite memory and run code on a server. Oracle $ORCL included a fix for the flaw in its November Critical Security Patch Update, which it released
[9]
AI Is Breaking the Entire Structure of Software That Powers the World
The systems we take for granted are crumbling before our eyes. AI's ability to churn out code and find vulnerabilities is making swiss cheese out of the world's digital infrastructure. On Tuesday, the cybersecurity firm Calif revealed that it used a new OpenAI model to discover a critical flaw in
[10]
AI finds security flaw in Java programming language
The company that found the PHP flaw in October has used AI to uncover another security flaw in the widely used Java programming language. The US-based security firm Calif, which was involved in uncovering the PHP flaws patched in October, has used AI to find another critical vulnerability, this
[11]
Major security bug found in Java, a software platform vital to the internet
A cybersecurity startup revealed Tuesday it had used a new artificial intelligence model to uncover a critical software bug in Java, a popular programming language that is used by government agencies and about 90% of Fortune 500 companies. It was the latest evidence of a reality facing companies
[12]
Java Security Flaw Discovered by AI: Why 64 'A' Characters Could Put Systems at Risk
From NASA to major corporations, Java4Shell vulnerability raises alarm over a potentially widespread cybersecurity threat. A string of 64 "A" characters may look harmless, but under specific conditions it can trigger a serious Java security flaw, potentially allowing hackers to take control of
[13]
Java vulnerability that can put millions of systems at risk discovered using AI, US issues warning
The US and South Korea have urged organisations to apply security updates. Soon after a major PHP2Shell flaw was discovered in the programming language PHP, flaws have now been discovered in Java that have drawn the attention of America's Cyber Defense Agency. Calif, a cybersecurity startup,
[14]
AI just found a critical Java flaw affecting banks and hospitals
AI gets a lot of flak these days, and to be fair, some of it is deserved. With all the AI slop flooding our social media feeds, it's easy to forget that the technology can be used for good. Case in point, security research. A cybersecurity startup called Calif says it used AI to uncover a critical
Share
Copy Link
Calif, a cybersecurity startup, has disclosed a critical flaw in Java, the nearly 30-year-old programming language that underpins much of the world's banking, healthcare and government software
4
8
. It says an AI model found the bug, and unpatched servers running the software are exposed to takeover4
.
Source: Digit
Java runs on about 90 percent of Fortune 500 companies' systems, per multiple outlets, and two outlets rank it the world's third most popular programming language
8
5
10
. That ubiquity, across about nine in ten big US companies, is what turns one bug into a systemic patching problem. "Java is basically critical infrastructure... a lot of big companies are going to have their work cut out to get it fixed," said Bruce Dang, a senior researcher at Calif2
.Calif's write-up locates the bug in the optimized Base64 decoder inside OpenJDK, the open-source core of Java
10
8
. A string of 64 'A' characters plus an equals sign makes the decoder write past its memory buffer; finessed, the overflow can run attacker code with administrator-level access10
8
3
. Base64 handles logins, file uploads and network messages, so the flawed path is easy to reach8
.
Source: Engadget
It affects OpenJDK 17.0.6 and newer, including Oracle Java SE, Amazon Corretto and Microsoft's build, but only on x86-64 chips with a certain vector feature. Machines with ARM chips are unaffected
8
7
. Calif says it showed it could run attacker code without any login on default installs of Keycloak, Atlassian's Jira and Confluence, JFrog Artifactory, Jenkins and PingFederate8
3
7
. A second set of products, including Apache Kafka and Elasticsearch, needed a login first, per its demos8
7
. Keycloak, a widely used open-source login system, guards organizations including NASA and the European physics lab CERN, per one outlet3
9
. No single exploit works everywhere, the company says; each target needs its own attack8
7
.Everyday personal devices are not at risk
1
. The exposure falls on enterprise and cloud systems1
. Most web traffic passes through firewalls and proxies that filter malformed data before it reaches Java, which reduces but does not remove the risk1
.Calif is a US security startup that uses leading AI models to hunt for serious software vulnerabilities
4
2
. Java4Shell is its second major find in weeks7
. Researcher Quang Luong at Calif used GPT-6 Cyber, an OpenAI model that has not been released publicly, to locate the flaw in one hour4
2
8
. He built a working exploit in two more, for under $200 in computing tokens4
8
. The New York Times says it independently reproduced the attack in under an hour4
.The company says it went after Java because many people were unfamiliar with PHP, the web programming language, a nod to its October find
7
. That framing drew pushback from a peer: Jan Schaumann of Akamai accused Calif of 'publicity-maxxing' on a security mailing list7
. In October the company used OpenAI and Google models to uncover two PHP flaws that had gone undetected for over a decade2
5
8
. That software is estimated to run some 70 percent of websites2
8
. AI researcher Toby Walsh characterized the Java bug as having hidden in plain view for some 12 years5
.Related Stories
Calif told Oracle a month before going public, then coordinated with the Five Eyes governments so fixes would be ready
4
2
6
. It also coordinated with vendors including Google, Microsoft, Amazon and Apple4
2
. Oracle released its critical patch update on Tuesday with the fix inside, though it did not name the bug2
8
. The US cybersecurity agency CISA urged immediate patching, starting with systems exposed to the internet5
1
.
Source: FT
The two sides disagree on severity. Calif scored the flaw 9.2 out of 10
8
. Oracle's advisory put it at 8.1 with high attack complexity8
7
. The scores come from different versions of the industry's vulnerability-scoring standard, so they are not directly comparable8
7
. For operators who cannot patch yet there are stopgaps: the agency recommends validating and limiting Base64 input, and Calif suggests launching OpenJDK with an option that switches off the faulty optimization8
.The trigger string was posted publicly on 9 September in an OpenJDK change proposal that was later abandoned, per Calif's account
7
. Attackers have therefore had the bug's technical core for two months7
. The security firm watchTowr reported seeing exploitation attempts against its sensors, though no attacks in the wild had been confirmed as of Tuesday evening8
7
. Calif estimates the bug could linger in unpatched systems for five to 10 years3
. Calif's proposed structural fix is what it calls vibe-uncoding: using AI to find and delete unneeded optimizations like the one that caused this bug7
. That code, it notes, helps few applications but ships to everyone7
.The disclosure landed amid a wave of US tech outages, from Microsoft to major banks, none of them linked to the Java bug
2
5
6
. Secretary of State Marco Rubio said the country is 'not under systemic attack' and that each event has a separate, non-hostile explanation2
5
.Calif's researchers describe the find as the arrival of the 'centaur hacker', a human working with AI. Chief executive Thai Duong said fully autonomous AI hacking may be near
9
. "I don't know how much time we have -- maybe three months, maybe three years," he said. "But right now, the best hacker is the centaur hacker."9
Summarized by
Navi
[1]
22 Oct 2026•Technology

18 Sept 2026•Technology

28 Jul 2026•Technology
