Oracle today announced new autonomous AI innovations for Oracle AI Database that will help customers build, deploy, and scale secure autonomous AI applications supporting large-scale production workloads. Oracle AI Database combines autonomous AI and data in operational databases and analytical lakehouses. It enables AI agents to securely access real-time business data where it resides and use it with LLMs trained on public data to deliver business insights. Customers can choose AI models, autonomous frameworks, open data formats, and deployment platforms. Additionally, customers using Oracle Exadata will be able to enjoy Exadata Powered AI Search, which delivers autonomous AI at the largest possible scale, with accelerated AI queries for high-volume, multi-step autonomous AI workloads.
“The next wave of enterprise AI will be defined by customers’ ability to use AI in critical production enterprise systems to securely deliver breakthrough innovations, insights, and productivity,” said Juan Loaiza, executive vice president of database technologies at Oracle. “With the help of Oracle AI Database, customers don’t just store data, they make it available to AI. By combining AI and data, we help customers quickly develop and manage autonomous AI applications that can query and act on real-time business data with the robustness expected by public companies in cloud and on-premises environments.”
Faster innovation with AI designed for data
With AI capabilities built for data, Oracle AI Database helps eliminate the need to create and maintain data movement pipelines that add complexity and security risks and could worsen outcomes. These new capabilities include:
• Oracle Autonomous AI Vector Database offers the simplicity of a vector database with all the power of Oracle AI Database. It helps developers and data scientists develop vector-based applications quickly and easily using intuitive APIs and a simple-to-use web interface. Based on Oracle Autonomous AI Database, it combines a simple developer experience with enterprise-class security, reliability, and scalability. Currently available on a limited basis, the Autonomous AI Vector Database can be accessed through Oracle Cloud Free Tier or in a reduced-price developer mode. With a single click, customers can easily upgrade to enjoy the full power of Oracle Autonomous AI Database as their needs grow, with full support for graph, spatial, JSON, relational, text, parallel SQL, etc. data, eliminating the need for separate databases and complex autonomous AI workflows between databases.
• Oracle AI Database Private Agent Factory enables business analysts and domain experts to create data-driven agents and workflows quickly and securely. AI Database Private Agent Factory is a no-code AI agent builder that runs as a container in public clouds or on-premises environments, ensuring the security of customers’ data by allowing them to create, deploy, and manage AI agents without the need to share data with third parties. AI Database Private Agent Factory includes multiple pre-built AI agents specialized in data, such as Database Knowledge Agent, Structured Data Analysis Agent, and Deep Data Research Agent. Other approaches rely on orchestrating external agents or must make calls to different types of databases. Oracle has simplified autonomous AI for enterprise users by including it in AI Database to deliver consistency and simplicity with enterprise-class security, resiliency, and scalability for all autonomous workloads.
• With Oracle Unified Memory Core, users can store context for AI agents in a single system. Enables unique low-latency reasoning for vector, JSON, graph, relational, spatial, and columnar data in a converged engine, with consistent transactions and security.
Minimize the risks associated with AI
Oracle AI Database helps customers securely store their data and protect it from external attacks, internal misuse, accidental disclosures, and unwanted exposure to LLM in multicloud, hybrid, and on-premises environments. These new capabilities include:
• Oracle Deep Data Security implements powerful end-to-end, user-based data access rules in the database. Each end user or AI agent acting on behalf of an end user can see only what they are authorized to see. You can apply sophisticated rules based on role and function. For example, you can specify which parts of a customer account salespeople, the finance team, warehouse staff, management, support agents, and family members are authorized to see. This provides unique end-user data security capabilities to protect against new AI-era threats, such as prompt injection, through native declarative database controls that enforce least privilege access principles. By centralizing and decoupling application code security, it helps customers easily determine who can see what data, continually update access rules as new threats emerge, and provide effective guardrails for agents working in the Oracle AI Database. Security at the source of the data, the database, provides superior protection when AI agents directly access data on behalf of end users.
• Oracle Private AI Services Container enables customers with strict security requirements to run AI model instances by preventing data from being shared with third-party AI providers or data being sent outside their firewall. Additionally, it helps mitigate performance bottlenecks by allowing customers to offload compute-intensive AI tasks (such as generating vector embeddings) outside of the database, helping to ensure the security of data in their environment. The container can be deployed in the public cloud, private clouds or on-premises, even in isolated environments.
• Oracle Trusted Answer Search offers businesses an accurate, verifiable, and deterministic way to use AI to provide answers to end users. Instead of directly using an LLM to answer an end-user question, Trusted Answer Search uses AI Vector Search to link the question to a previously created report. This helps mitigate the risk of hallucinations from probabilistic LLMs or misinterpretation of queries by them.
End dependency through open standards and frameworks
Oracle AI Database runs on all leading cloud providers and in hybrid or on-premises environments. It gives customers flexibility to choose the AI model and application-level autonomous AI framework that best fits their needs. They can create, deploy and run autonomous AI applications using open standards and data formats. These new capabilities include:
• Oracle Vectors on Ice provides customers with native support for vector data stored in Apache Iceberg tables. AI Vector Search can read vector data directly into Iceberg tables, create vector indexes to speed up vector search, and automatically update those indexes when the underlying vector data changes. Oracle Vectors on Ice enables AI search across data in data lakes and unified search across enterprise data in the database and vectors stored in a data lake. This helps customers achieve unified intelligence across all their databases and data lakes.
Oracle AI Database runs on all leading cloud providers and in hybrid or on-premises environments
• Oracle Autonomous AI Database MCP Server enables external AI agents and MCP clients to securely access the Autonomous AI Database and its capabilities without requiring custom integration code or manual security configurations. It acts as a plugin for Oracle SQLcl MCP Server for Oracle AI Database, available through the Oracle SQL Developer VS Code extension.
“In the era of autonomous AI, a unified memory core is essential so that agents can maintain context across different types of data, whether vector, JSON, graph, columnar, spatial, text or relational, without the latency or rigidity of external synchronization,” said Steven Dickens, CEO and principal analyst at HyperFRAME Research. “Only Oracle AI Database delivers this in a single mission-critical engine with simultaneous transactional and analytical processing, high availability, and ironclad security, enabling real-time reasoning with dynamic business data. Organizations without this foundation will struggle with fragmented and unreliable agents, while those that turn to Oracle will gain a decisive advantage in deploying scalable AI.”
Customers and developers can take advantage of Oracle AI Database’s new autonomous AI capabilities now to start developing and deploying breakthrough autonomous AI applications without moving data, acquiring new skills, or facing obstacles due to lack of database scalability and security barriers to autonomous AI.
