Oracle AI World Preview: Database Modernization as Foundation for Enterprise AI

Historically, organizations have pursued database modernization primarily to achieve operational savings and efficiency gains. According to theCUBE Research, however, a more substantial opportunity has emerged: enterprises that build AI initiatives atop a modernized data foundation can unlock considerably greater value.

Moving past a scattered data landscape—which impedes team velocity and complicates data stewardship—represents a critical objective. Successful AI deployments demand repeatability and scalability in live environments, themes that will dominate discussions at Oracle AI World later this month in Las Vegas.

For Oracle customers, the strategic question is becoming: Are we simply reducing infrastructure cost, or are we creating a better platform for repeatable AI execution? The value is not only in automation, standardization and operational efficiency. The bigger opportunity is using that foundation to get onto the AI learning curve earlier, free skilled teams from routine administration, improve governance consistency and accelerate the development of AI use cases over time. Database modernization is still an IT decision. But, increasingly, it is also an AI readiness decision that impacts all parts of the business.

Dave Vellante, theCUBE Research

TheCUBE, the livestreaming division of SiliconANGLE Media, will deliver on-site reporting and exclusive conversations with Oracle leadership and industry figures on October 28. Coverage will examine how Oracle is embedding agentic AI into business processes, linking dependable data sources with intelligent automation, and tackling infrastructure, security and governance demands as AI scales across enterprises.

Database modernization to ensure trust

Oracle AI World will showcase progress across several critical product domains, with particular emphasis on implementing safeguards and establishing confidence in AI systems.

For enterprises to rely on AI-driven outputs, those outputs must demonstrate accuracy, security and dependability. This requirement has grown urgent as autonomous agents construct applications and execute decisions, according to Juan Loaiza, executive vice president of mission-critical database technologies at Oracle.

Guardrails don't cut it. You can't simply drop AI on top of an architecture that was never designed for these challenges AI brings and expect it to work. What we're doing is different. We're ensuring trust across the entire technology stack. The only way we think is possible to build AI solutions that we can trust is to architect the whole thing together — AI, data and AppDev all in one whole.

Juan Loaiza, Oracle

Oracle unveiled a three-part approach earlier in the year to counter security risks introduced by expanded AI adoption. Since AI systems have opened new vectors to sensitive corporate information, Oracle has rolled out tools targeting database security, patch management, testing and operational lifecycle oversight.

There's so much optimism on AI and how it can improve productivity, but there's also a lot of concern when it comes to cyberattacks. One of the key focus areas for us in Las Vegas is to really show what we're doing in the product, how we're designing features to really start at the data layer and go all the way to the app layer to help customers move from AI prototypes to trusted AI apps at scale.

Arpan Shah, Oracle

Enhancements for Lakehouse and AI Database technologies

A second major theme centers on intelligent data infrastructure. The Autonomous AI Lakehouse merges Oracle's Autonomous Database with Apache Iceberg's open table format, establishing a unified analytics platform accessible across all major cloud providers.

Oracle plans to introduce multiple new features within Autonomous AI Lakehouse, as noted by Nipun Agarwal, SVP of Oracle Autonomous AI Lakehouse.

The lakehouse is transforming from being an analytic platform into a decision backbone for enterprise AI. What that means is basically that in the past, customers would come to a lakehouse for running their analytic workloads. But, now, lakehouse is the right, appropriate platform both for analytic as well as AI workloads.

Nipun Agarwal, Oracle

The Autonomous AI Database delivers a fully automated, cloud-native data system leveraging machine learning and AI to handle management, optimization, patching and protection autonomously. Its cross-cloud reach permits organizations to deploy on Oracle Cloud Infrastructure or through Oracle-managed services across Amazon Web Services, Microsoft Azure and Google Cloud. Supporting multimodal data handling, the platform manages both transactional and analytical workloads while streamlining and automating database administration.

We'll be talking a lot about building and integrating AI agents with Autonomous AI Database at AI world, of course. But if you're only looking at how LLMs interact with enterprise data from an application standpoint, you're missing a big part of the picture. An often overlooked ingredient is that AI applications need to be built on the right data foundation. And that is where the real strengths of Autonomous AI Database come in.

Çetin Özbütün, Oracle

New platform capabilities

Oracle has strengthened its platform offerings this year through additions including APEXlang, a human-readable language for defining applications that gives coding agents a structured target while Oracle handles runtime, security and lifecycle administration.

What APEXlang does is it takes all this configuration or all these properties that you define with a low-code system traditionally through a user interface, but it's captured in a formal language with a formal grammar and that can be compiled. So, you can use a coding agent like Codex or Claude, and you just simply say, 'Build me an application to do something.' And it will generate this language.

Mike Hichwa, Oracle

Oracle has been progressively broadening its high-availability Exadata offering to address AI-specific and location-dependent scenarios. Previously, migrating database virtual machines using Remote Direct Memory Access on Exadata presented insurmountable technical barriers.

In June, Oracle introduced Exadata VM Live Migration, enabling an active Oracle Database to transition between servers without service interruption while operating on Exadata. Oracle AI World will provide deeper detail on this capability, according to Jia Shi, group vice president of software development at Oracle.

Before this June, nobody could actually move their database VM running RDMA from one host to another in a live fashion. So, we are really happy to announce that we finally solved that technical challenge. This is truly unique in the sense that we're able to launch the capability on Exadata where customers can move their database VM running an active database workload with all the RDMA connections intact from one server to another.

Jia Shi, Oracle

How to follow the event

TheCUBE will livestream its Oracle AI World coverage on October 28, with on-demand access to all interviews and reporting available afterward. Viewers can tune in through theCUBE's website, YouTube channel or via SiliconANGLE's broader event coverage.

SiliconANGLE produces two regular programs featuring enterprise technology analysis. "TheCUBE Pod," hosted by John Furrier and Dave Vellante, examines major industry trends including AI, cloud, cybersecurity and infrastructure, available on Apple Podcasts, Spotify and YouTube. The weekly "Breaking Analysis" program, also hosted by Vellante and drawing on theCUBE interviews and Enterprise Technology Research spending data, streams on the same platforms.