Oracle Unveils Deep Trust AI Architecture

The four-layer framework was unveiled at Oracle AI World in Las Vegas on October 26, alongside its APEX AI Application Generator.

Updated on Nov. 5, 2026 in Artificial Intelligence

Oracle Unveils Deep Trust AI Architecture

Live Poll

Do you trust AI-generated software to handle your personal data safely?

On October 26, Oracle unveiled its Deep Trust AI Architecture at Oracle AI World in Las Vegas. The framework spans four layers: software platform, data, data access, and application development.

Why it matters

The architecture pairs AI-driven application generation with built-in permission and business-rule controls, which could affect how enterprises build and govern AI-enabled software. Its emphasis on open, human-readable definitions and centralized runtimes places Oracle's approach within the broader push to make enterprise AI development more auditable.

The architecture is organized into four layers spanning software platform, data, data access, and application development. The extraction does not provide benchmark performance figures or independent testing results for these components.

The players

Oracle

Oracle is a multinational enterprise software and cloud company best known for its database products and enterprise applications.

The details

The APEX AI Application Generator converts natural-language descriptions into APEXlang definitions, which Oracle APEX then runs in a centralized runtime. The generator produces an application definition rather than executable application code, and Deep Data Security applies end-user permissions when Oracle AI Database processes SQL requests.

May: Oracle introduced APEXlang as an open, human-readable application definition language.
October 6: Oracle announced general availability for APEX 26.2 on-premises deployment.
October 26: Oracle unveiled the Deep Trust AI Architecture at Oracle AI World in Las Vegas.

The Tech Race

The Deep Trust architecture extends Oracle APEX, the company's existing low-code application platform, by layering AI-driven generation on top of APEXlang definitions. It positions Oracle to compete with other enterprise vendors racing to embed generative AI directly into application development tools.

Developers who build internal business applications could describe them in natural language and have APEX generate the underlying definitions, reducing manual coding. Organizations would still need to review generated definitions and confirm that Deep Data Security and SQL Assertions enforce their own access and business rules.

The takeaway

Generated application definitions still require human review, so organizations adopting AI-driven app generation should verify permissions and business rules before deployment. Checking that access controls apply at the database level is a practical first step.