Oracle recently launched its APEX AI Application Generator on Autonomous AI Database Serverless. The new capabilities address a problem that AI coding tools have created for nearly every enterprise, as users now generate working software from prompts, often without IT involvement.

Note: The author and his firm, NAND Research , have in the past provided advisory and other analyst services to Oracle.

Gartner forecasts that developers outside formal IT departments will account for at least 80% of the user base for low-code tools by 2026, up from 60% in 2021. Agentic AI has accelerated this shift.

The productivity gains are real, but AI-generated applications can leave behind substantial amounts of code that must be secured, tested, patched, and maintained for years.

When the employee who prompted the application changes roles, or the model that wrote it is replaced, the organization can inherit code that few people understand. This can become an audit and security risk that compounds with every new application.

Oracle’s APEX AI Application Generator changes what the AI produces from the start. Its generator outputs structured application definitions that existing review, testing, and deployment processes can govern, and the managed APEX runtime handles execution and patching. That design choice has implications well beyond Oracle’s installed base.

How Oracle Governs What AI Generates

The generator builds on APEX, the low-code platform that has been included with Oracle Database for more than two decades. Its centerpiece is APEXlang, a new declarative language that describes an application, including pages, navigation, forms, reports, dashboards, and business logic, in human-readable text.

AI agents write and edit APEXlang definitions, which are compiled into APEX application metadata. The managed APEX runtime then executes the resulting applications. Oracle compares the model to SQL, where a developer uses SQL to specify the desired result from the data, while the underlying database determines how to deliver it.

APEXlang applies the same division of labor to applications. While the AI expresses intent, the platform handles execution, enforces security, and manages upgrades.

Because APEXlang follows a published grammar, AI agents operate within a narrow, validated set of instructions, shrinking the attack surface of generated applications.

The architecture translates into five business outcomes:

  • Auditability: Every AI-generated change appears as a readable text diff that flows through the existing source control, code review, testing, and rollback processes.
  • Lower maintenance burden: Applications on the managed runtime pick up platform patches and security updates without being recompiled or rewritten.
  • Model choice: Oracle publishes the APEXlang grammar and supporting AI skills openly, enabling organizations to use third-party coding agents, such as OpenAI’s Codex, alongside Oracle’s App Builder and the VS Code tooling.
  • Inherited data security, scalability, and availability: Generated applications inherit Oracle AI Database’s access controls, including Deep Data Security, dynamic scalability, and high availability.
  • No additional licensing cost: APEX, including the APEX AI Application generator, is included at no additional cost with supported Oracle AI Database editions. Inference charges from the selected AI provider remain the customer’s responsibility.

Oracle applies the same constraint-based philosophy across its applications. APEX 26.1 introduces AI Interactive Reports, which convert natural-language requests such as “pivot sales by region” into standard report settings without generating AI SQL.

APEX 26.1 lets developers embed conversational AI agents grounded in developer-provided tools, constraining agent actions to approved capabilities and reducing the risk of unauthorized operations.

The Competitive Landscape

The APEX AI Application Generator enters a crowded market. Enterprise low-code vendors have integrated generative AI into their platforms, and general-purpose coding agents such as Codex, Claude Code, and Cursor now generate full-stack applications from a written specification.

Microsoft Power Apps with Copilot, Salesforce with Agentforce, ServiceNow App Engine, OutSystems, and Mendix all compete for the same internal-application budget.

While Oracle APEX shares the metadata-driven approach of platforms such as Salesforce, OutSystems, and Mendix, it differentiates itself through deep integration with Oracle AI Database, where many enterprises run their most mission-critical applications and store sensitive data.

By enforcing security controls directly in the database, Oracle APEX protects data at the source, not just at the application layer, while providing a foundation for secure, governed, and scalable AI-generated applications.

Oracle publishes its definition language so any AI model can target it. It also carries no additional license fee for organizations already subscribed to Oracle AI Database.

The limits are equally clear. Every advantage depends on Oracle AI Database, which has an extensive global installed base, leaving the generator out of reach for organizations standardizing on PostgreSQL or Microsoft’s data stack. Power Apps reaches far more business users via Microsoft 365 and Teams.

Buyers evaluating this category should weigh three factors: where the data resides, who owns and maintains the artifact throughout its lifetime, and how easily the organization can switch AI models as capabilities improve.

Analyst Take: What Comes Next

Over the next 12 to 24 months, the industry conversation will shift from how fast AI can produce an application to who is accountable for it once it runs in production. Auditors, regulators, and security teams already want to know how organizations control software no human wrote line by line.

Platforms that clearly show what an AI changed and keep generated applications patched without human rework will gain an advantage in that discussion.

For the many enterprises already running core operations on Oracle, the calculus is straightforward. The APEX AI Application Generator lets AI build business applications while IT retains control over what reaches production, with no additional licensing costs.

As AI transforms how enterprise software is built, the next competitive frontier won’t be defined simply by how fast applications can be generated, but by how securely they can operate, how effectively they can be governed, and how confidently they can evolve.

Oracle APEX AI Application Generator marks a fundamental shift in that equation, combining the speed of AI-driven development with the security, scalability, and governance of Oracle AI Database.

Ultimately, the future of enterprise software isn’t just about what AI can build. It’s about what enterprises can trust AI to build.