Oracle launches new platform to streamline AI development

A new platform for AI

With AI Data Platform, Oracle is the latest data management vendor to introduce a unified feature set for building and managing agents and other AI applications.

This is a natural extension of many existing Oracle elements to help organizations build agentic applications on top of their Oracle infrastructure. ... By tying the pieces together, Oracle reduces the need [for customers] to adopt third-party alternatives.

Data management vendors began adding AI development capabilities following OpenAI's November 2022 launch of ChatGPT, which represented a significant improvement in generative AI (GenAI) technology and led many enterprises to increase their investments in building AI applications.

Over the past 18 months, agents have become the focus of most AI initiatives. Unlike GenAI chatbots that require human prompts to act, agents are programmed with contextual awareness and reasoning capabilities that enable them to autonomously take on certain tasks.

Databricks and Snowflake each unveiled agentic AI development capabilities in the spring, including support for MCP. More recently, Microsoft and Teradata were among many other vendors to add features aimed at improving their agentic AI development.

Now, Oracle is doing the same with AI Data Platform, which was designed to enable the following, according to the vendor:

• Simplify turning raw data into insight-generating applications by unifying data lakehouse and AI capabilities in one platform.
• Provide a single environment for data engineers, data scientists and AI developers to foster collaboration and speed development of AI tools.
• Build agents that autonomously orchestrate workflows, trigger alerts and improve overall business efficiency.
• Provide the scale, performance and trust that enterprises require to not only develop agentic AI applications but also put them into production.

Beyond the general-purpose version of AI Data Platform, Oracle plans to provide industry-specific versions with prebuilt capabilities that are tailored to enterprises in healthcare, financial services, construction and consumer goods. In addition, Oracle is planning versions of the platform for its Fusion applications and its NetSuite ERP tools.

Stephen Catanzano, an analyst at Enterprise Strategy Group, now part of Omdia, called AI Data Platform "highly significant" given that it provides a unified feature set for creating agents.

"The platform addresses the critical challenge of connecting generative AI models securely with enterprise data, applications and workflows in one comprehensive solution," he said. "The advantage of packaging [previously separate capabilities] together is creating a single, enterprise-grade platform that simplifies the entire AI lifecycle."

Perhaps AI Data Platform's most significant capability is the Agent Hub, Catanzano continued.

"The Agent Hub stands out as most significant because it abstracts the complexity of navigating multiple agents, interprets requests, invokes the right agents, presents recommendations and enables immediate action," he said. "This capability transforms AI from a technical tool into an accessible business solution."

However, while a valuable suite for Oracle users, AI Data Platform is more about bringing together existing capabilities than adding new ones, according to Petrie. There are additions within those existing capabilities, such as AI Database 26ai to better enable vector search and use unstructured data in conjunction with structured data to train AI tools, but unification is the platform's primary significance to him.

"The technology itself strikes me as incremental improvements on existing tools," Petrie said. "But taken together, the AI Data Platform strengthens Oracle's role in AI innovation. They come at it from a position of strength, [and] now they're adding the ability to vectorize unstructured objects and assemble knowledge graphs, which diversifies model inputs and enriches outputs."

Competitive standing

While AI Data Platform is a significant addition for Oracle, it is not a suite that will differentiate the vendor from its peers, according to Petrie.

Oracle has long competed with fellow tech giants AWS, Google Cloud and Microsoft. In addition, Databricks and Snowflake have emerged as significant competitors in recent years. Each of those rivals already provides an AI development environment in addition to its data management capabilities. For example, Microsoft Fabric features OneLake to unify data management, data discovery capabilities, data and AI governance, and AI development.

However, with the launch of AI Data Platform, Oracle is now able to keep customers from potentially defecting, Petrie said.

"I don't see a lot of differentiation compared to Databricks, Snowflake or the hyperscalers," he said. "But this definitely reduces the ability of those competitors to steal share from Oracle."

And that is significant, according to Catanzano, who noted that the new platform's deep integration with other Oracle platforms may not be a differentiator for potential new customers, but it is for Oracle users.

"The platform is a strong differentiator, particularly through its deep integration with Oracle's application ecosystem -- Fusion, NetSuite -- and the promise of prebuilt integrations for major Oracle application suites," Catanzano said. "The capabilities for seamless connection to critical business application data across [Fusion] provide unique value for existing Oracle customers."

Eric Avidon is a senior news writer for Informa TechTarget and a journalist with more than 25 years of experience. He covers analytics and data management.