Why does anyone need to learn C++, Python or Java today if they can vibe-code their way to a piece of software without knowing a programming language? Heck, someone even used AI to create a suite of design software similar to Adobe’s costly tools like Photoshop and Lightroom.
Ask Scott Sellers, president and chief executive officer of Java software vendor Azul Systems, however, and he will tell you the ubiquitous programming language used in servers, cars and all manners of devices today is going to be great in the era of AI.
Java, he argues, benefits from vibe coding because its inherently verbose (or long-winded) code won’t be as much of an issue to programmers when AI can help manage that.
“People don’t love Java for succinctness,” he said. “One line in Python might take 20 lines of Java but if a code generation tool is doing it for you, why do you care?”
Just as importantly, Java’s resilience, security and performance remain the reasons why people will continue to use the programming language, he told Techgoondu in an interview in Singapore last week.
Trusted to run the most mission critical servers for big corporations over the years, Java has been central to most of today’s digital infrastructure.
On a cloud server, it offers fast multi-threaded performance to manage concurrent tasks. It provides near-native execution speeds close to C++ and the managed memory safety and security of languages like Python.
In particular, new versions in recent years (it’s at Java 27 now) support virtual threads, which Sellers says offers “dramatic scaling” in performance.
Plus, Java updates are more frequent now with a new release every six months, which has resolved an old bugbear for many developers, he added.
The servers that run many of the enterprise systems now powering up AI agents are based on Java, he pointed out, and as more of them come online to support the explosion in AI usage, Java will remain key to the equation.
What drives many companies to buy a version of Java from Azul – as opposed to the more expensive one from Oracle – is to make sure they are running their digital infrastructure securely.
Azul also offers tools to take stock of the various versions of Java running in an enterprise, not just to avoid dreaded Oracle Java bills but also to ensure the software is up to date and patched.
In some instances, businesses can have third-party software that are installed and using Java that they do not know about. That is a loophole waiting to be exploited by AI-powered hackers.
“The world is a different place now and the frontier AI models have shown that the guardrails in place are often not sufficient,” said Sellers.
Identifying where the vulnerabilities are is the first step for many businesses, he noted, adding that Azul provides patches even for old Java versions so they can run securely until the apps that depend on them can be updated to run on more modern versions of Java in future.
Optimisation at a time of heightened costs is another idea that Azul wants to raise with businesses.
The company sells a version of Java, called Prime, which promises to run much faster than regular plain vanilla Java. This, it says, can cut down cloud server costs and push more money into AI budgets.
Many companies are still yet to understand that running a more optimised version of Java can reduce the load on cloud servers which brings savings that can be used for AI, said Sellers.
“The demand for the AI budget is insatiable,” he noted, “and many enterprises are (only) going through the early stages of AI adoption.”
