-
Oracle debuted an AI-powered coding tool
-
The tool is designed especially for Java and SQL
-
Oracle's tool is expected to garner strong adoption among enterprises, IDC said
Oracle is leaning into its Java roots with a new artificial intelligence-powered code assistant which is designed to make it easier for developers to refactor and upgrade old code.
The company joined the ranks of Amazon, Google Cloud and GitHub, unveiling its own Oracle Code Assist tool. Oracle’s version, of course, is tailored for the Java and SQL programming languages. The former is almost a no brainer considering Oracle owns the official version of Java, having acquired it when it bought up Sun Microsystems back in 2010.
In addition to surfacing recommendations based on enterprise-specific best practices and codebases, the tool will provide developers with code snippets, dependency analysis, error mitigation alternatives, test cases, annotation, summarization and documentation.
Gartner recently predicted that three-quarters of enterprise software engineers will use AI code assistants by 2028, compared to around 10% at the start of 2023. The firm added that 63% of enterprises had already either deployed or were testing such tools as of the end of last year.
Why? Well, beyond just saving time and cutting costs, these tools can help improve code quality and maintainability, speed go-to-market times, cut down on bugs and reduce technical debt and boost developer retention, according to Gartner.
Indeed, Wayfair’s Head of Martech, Data and Machine Learning Matt Ferrari told us last year that as AI tools increasingly tackle code basics, developers will be able to start thinking differently and focus more on things like value-added features and the customer experience.
Oracle is facing off with some powerful competitors in the AI code assistant market, including Amazon (which has Amazon Q), Google Cloud’s Gemini Code Assist and CodeGemma and GitHub’s Copilot Enterprise.
However, IDC Research VP Arnal Dayaratna noted in a statement Oracle’s tool is “likely to achieve widespread adoption by prominent organizations globally” given “the ubiquity of Java and databases in the enterprise.”
"Oracle’s access to both Java and SQL-related training data means that code suggestions that are generated by Oracle Code Assist are likely to be more accurate and insightful than analogous suggestions delivered by its competitors," he added in an email to Fierce. "Additionally, the technology’s enablement of automated upgrades of Java from one version to another has the potential to help organizations upgrade to newer versions of Java more frequently, and thereby reap the benefits of the six month release cadence for Java SE."
Update 5/7/2024 2:30 pm ET: This story has been updated to add additional commentary from Dayaratna.