TLDR: Oracle has announced the general availability of Oracle AI Database 26ai, a new AI-native database that embeds artificial intelligence capabilities directly into its core architecture. This long-term support release, replacing Oracle Database 23ai, aims to simplify and accelerate AI adoption by enabling AI training and inference across multicloud and on-premises environments without requiring data movement or third-party tools. Key features include AI Vector Search, Model Context Protocol (MCP) server support, and a framework for deploying AI agents within the database.
Oracle Corporation officially unveiled its new flagship offering, Oracle AI Database 26ai, at Oracle AI World in Las Vegas on October 14, 2025. This significant release marks a strategic shift, architecting AI directly into the core of data management to support artificial intelligence training and inference seamlessly across diverse environments, including Oracle Cloud Infrastructure (OCI), Amazon Web Services, Microsoft Azure, Google Cloud, and on-premises systems like Oracle Exadata and Oracle Database Appliance. The new database is also available in Oracle Database Free for developers.
Oracle AI Database 26ai is a long-term support release designed to supersede Oracle Database 23ai. Existing customers can transition to 26ai by applying the October 2025 release update, with no requirement for a full database upgrade or application re-certification. Advanced AI features, such as AI Vector Search, are included at no additional charge.
Juan Loaiza, executive vice president of Oracle Database Technologies, emphasized the simplicity and utility of the new approach, stating, “By architecting AI and data together, Oracle AI Database makes ‘AI for Data’ simple to learn and simple to use. We enable our customers to easily deliver trusted AI insights, innovations, and productivity for all their data, everywhere, including both operational systems and analytic data lakes.”
The ‘AI for Data’ vision is realized through a comprehensive integration of AI across the entire data and development stack. This includes AI Vector Search, AI for Database Management, AI for Data Development, AI for Application Development, and AI for Analytics. The database empowers customers to execute dynamic agentic AI workflows, combining private database data with public information to generate sophisticated answers and actions.
Key capabilities and integrations of Oracle AI Database 26ai include:
AI Vector Search: A native feature for efficient similarity searches.
Model Context Protocol (MCP) Server Support: Facilitates the integration of large language models (LLMs) and popular agentic AI frameworks.
AI Agent Framework: Enables the creation and deployment of AI agents directly within the database, supporting multi-step AI reasoning.
Open Standards Support: Compatibility with the Apache Iceberg open table format for data lakes and Open Neural Network Exchange (ONNX) embedding models.
Enhanced Security: Features quantum-safe encryption for data-at-rest and data-in-flight, utilizing government-approved algorithms to protect against future quantum computing threats. It also includes SQL Firewall for protection against unauthorized SQL activity and injection attacks.
Performance Acceleration: Leverages Oracle Exadata for AI, a high-performance database platform, along with remote direct memory access and tiered storage.
Private AI Services Container: Allows AI model instances to run on a customer’s own infrastructure, ensuring data privacy.
In conjunction with the AI Database 26ai, Oracle also launched the Oracle Autonomous AI Lakehouse, which supports the Apache Iceberg open table format, extending AI and analytics capabilities across enterprise data. The Autonomous AI Lakehouse is available on all major hyperscaler clouds and offers Exadata-powered performance with pay-per-use serverless scaling.
Holger Mueller, vice president and principal analyst at Constellation Research, commented on the release: “With Oracle AI Database 26ai, customers get both. It’s the single place where their business data lives—current, consistent, and secure. And it’s the best place to use AI on that data without moving it.” He further highlighted the inclusion of Agentic AI, enabling customers to build, deploy, and manage in-database AI agents using a no-code visual platform.
Also Read:
- Oracle Bolsters AI Agent Studio for Fusion Applications with New Marketplace and Broadened LLM Integration
- Informatica Enhances Oracle Cloud Offerings to Accelerate Enterprise AI Development
Future innovations planned for Oracle AI Database 26ai include a unified data model for relational, JSON, and graph data, data annotations for improved AI model output, a globally distributed database model, True Cache for automatic transactional data caching, and further support for Nvidia Corp.-based acceleration. Oracle also announced the AI Factory, a set of support services to assist customers in implementing AI solutions more efficiently.
