For a 30-year-old language, Java is having a remarkably turbulent year.
Azul’s 2026 State of Java Survey and Report, released this month, surveyed more than 2,000 Java professionals across five continents between September and November 2025. The findings point to a transition – AI-assisted coding has become the norm, cloud waste persists and enterprises are accelerating their move away from Oracle Java.
Azul, which sells OpenJDK-based alternatives to Oracle Java, commissioned the research, which was administered by Dimensional Research.
AI reshapes development workflows
Every respondent (100%) now uses AI code-generation tools, with 30% reporting that more than half their new code is created entirely by AI assistants. This complete adoption rate represents dramatic acceleration from 2024, when 76% of respondents were using or planning to use AI assistants.
ChatGPT-based tools lead adoption (52%), followed by Google Gemini Code Assist (43%), Microsoft Visual Studio IntelliCode (37%) and GitHub Copilot (35%).
While Python dominates AI model development, Java increasingly powers production AI services. The survey found 62% of respondents use Java to code AI functionality, up from 50% in 2025. Popular Java AI libraries include JavaML (45%), Deep Java Library (33%) and OpenCL (25%).
Beyond writing code with AI, developers embed AI capabilities into applications. The survey found 31% of organizations report more than half their Java applications include AI functionality.
Looking ahead, developers prioritize long-term support for modern versions (35%), built-in security features (34%) and runtime observability insights (32%) as critical capabilities for Java to remain competitive.
Java drives cloud costs and optimization efforts
Java’s enterprise ubiquity translates to substantial cloud impact. Among respondents, 43% report JVM (Java Virtual Machine)-based workloads account for more than half their total cloud compute costs.
In response, 97% have taken action to reduce public cloud expenses. Top strategies include using high-performance Java platforms (41%) — a category in which Azul competes — establishing usage KPIs (40%), leveraging cloud provider tools (38%) and using more efficient compute instances (37%).
Notably, organizations using high-performance Java platforms prioritize performance over cost. Primary motivations are improving application performance (61%), enhancing customer experiences (55%) and improving scalability (52%). Lower cloud costs (42%) are a beneficial side effect.
The report also exposed significant waste: 74% of respondents have more than 20% unused cloud compute capacity, suggesting widespread overprovisioning.
DevOps productivity faces persistent challenges
The report identifies significant productivity drains. Dead and unused code measurably affect 63% of DevOps teams, creating maintenance burdens as teams work around legacy code they can’t confidently remove.
Security vulnerabilities have intensified sharply: 56% of enterprises deal with common vulnerabilities and exposures in Java workloads daily (20%) or weekly (36%), up from 41% in 2025.
False positives compound the problem, with 30% of organizations reporting more than half their DevOps time wasted on security alerts that don’t represent actual production threats.
To address these issues, developers turn to automated tools: ChatGPT (58%), Gemini AI (51%), Amazon Q (32%) and Claude AI (31%).
Oracle Java migration
Some 92% of respondents expressed concern about Oracle’s licensing costs, up from 86% the previous year. In response, 81% have migrated, are migrating or plan to migrate some or all of their Oracle Java to open-source alternatives.
Cost remains the primary driver (37%), followed by preference for open source (31%) and uncertainty around Oracle’s policy changes (29%).
Among completed migrations, 84% said the process was easier than expected or went as planned. Additionally, 72% finished within one year, with 44% completing in six months or less, challenging perceptions that Oracle Java is uniquely complex to replace.
According to Scott Sellers, cofounder and CEO at Azul, the report “shows a community that is evolving quickly, embracing open technologies, accelerating cloud optimization and removing the friction that slows DevOps productivity.”