Oracle is expanding how employees interact with its enterprise software, introducing AI capabilities that connect business applications, workflows and data across its portfolio.
At Oracle AI World in Las Vegas on October 27, the company announced Oracle AI Workspace, one of four offerings from an expanded partnership with OpenAI. Through plugins in ChatGPT, employees would be able to access information, complete approved business workflows and take action across connected Oracle systems.
For ERP teams, the announcement raises a practical issue: how much business work can move into a conversational interface while permissions, approvals and audit requirements stay intact.
What AI Workspace Is Designed to Do
Oracle describes AI Workspace as a single user experience that combines OpenAI’s frontier models with Oracle applications, enterprise data and cloud services. It is designed to connect Oracle Fusion Applications, NetSuite, Oracle AI Database and Oracle Cloud Infrastructure (OCI), along with other enterprise tools and data sources.
Employees would describe a business goal, and plugins in ChatGPT would bring together the Oracle tools and information needed to accomplish it. Oracle says the workspace will maintain the governance, approvals and controls required for enterprise operations.
The wider partnership also introduces Oracle AI Security Mesh, a managed security-testing service using OpenAI cyber models, and Oracle App Composer, which uses Codex to help employees build applications while Oracle manages deployment and operations on OCI.
For smaller companies, Oracle announced the Oracle AI SMB Bundle, a free-to-start package combining ChatGPT Business, AI Workspace, Codex, App Composer, OCI and NetSuite QuickStart for core operations such as invoicing and cash flow. Eligible U.S. businesses can receive $12,000 in OCI credits annually for up to three years, subject to business verification, renewal requirements and individual product restrictions.
How Fusion Companion Fits Into Oracle’s AI Plans
AI Workspace arrived a day after Oracle introduced Fusion Companion, an intelligent layer designed to help employees execute work in Fusion Applications through supported AI tools such as ChatGPT.
Fusion Companion uses enterprise business context, data and user permissions to understand an employee’s objective and coordinate the transactions, workflows, approvals, AI agents and agentic applications needed to complete it. Users would be able to access those capabilities through ChatGPT, customer- or partner-built AI interfaces, or Fusion Agentic Home, the native AI experience within Fusion Applications.
The two announcements establish different roles. AI Workspace is designed to give ChatGPT users access to applications and enterprise data across Oracle’s portfolio, while Fusion Companion coordinates execution within Fusion Applications.
Companion is designed to coordinate Oracle’s agentic applications, including the 25 Fusion Claw-powered applications Oracle said on October 26 are now available. One of them, Ledger, works across large volumes of accounting entries to improve close readiness, investigate anomalies and resolve exceptions through authorized actions.
Enterprise Controls Move Into the AI Workflow
Letting an AI assistant act in finance or procurement systems carries more responsibility than using it to retrieve information. Those actions can affect payments, accounting entries, vendor records and other business data subject to internal controls.
Oracle says Fusion Companion is designed to operate within established identity, permissions, policies and approval controls, keeping enterprise execution and accountability within Fusion Applications.
That design puts weight on how organizations configure access and delegate authority. An AI-initiated finance transaction still needs a clear record of who requested it, which authority applied and who approved it. ERP teams will also need to determine whether an agent can combine individually authorized actions in ways that violate segregation-of-duties policies.
Organizations will need to establish which business information can pass through the external AI service, how permissions are enforced across connected systems and where the resulting activity is logged for audit purposes.
Snowflake Comes to OCI
In a separate October 27 announcement, Oracle and Snowflake said the Snowflake AI Data Cloud will be offered as a native, first-party service on OCI. Customers would be able to purchase, provision and consume Snowflake directly through Oracle under a single commercial relationship, using Oracle Universal Credits toward the service.
The planned offering supports enterprise data sources including Fusion Applications, NetSuite, Oracle Autonomous AI Database, MySQL HeatWave and Oracle NoSQL Database. It would let customers apply Snowflake’s analytics and AI capabilities to Oracle business data while maintaining existing governance requirements.
Oracle says supported native data integrations are designed to reduce reliance on custom pipelines, while keeping more data movement within OCI could lower cross-cloud egress costs. The potential savings will depend on customers’ existing architecture, integration requirements and data transfer patterns. Oracle has not announced a general-availability date for the offering.
For Oracle customers, the announcements open two areas for evaluation. AI Workspace and Fusion Companion could change how employees initiate and complete ERP work, while Snowflake on OCI could expand how organizations analyze the data those systems generate. Whether these capabilities deliver practical value will depend on their availability, integration requirements and the ability to preserve enterprise controls as more work moves through AI interfaces.
What This Means for ERP Insiders
Identify which ERP processes are ready for AI-driven execution. Oracle’s announcements create opportunities to automate work across finance, procurement and other business functions, but organizations need to distinguish routine tasks from transactions requiring human judgment. Start with processes that have defined rules, reliable data and established approval paths. Measure whether AI reduces processing time and manual intervention without introducing errors, weakening financial controls or creating additional reconciliation work.
Clarify how Oracle’s AI products fit into existing ERP investments. AI Workspace and Fusion Companion introduce different ways to access and execute business workflows, but their integration and licensing requirements remain unclear. Oracle customers should establish how these capabilities interact with existing Fusion functionality, embedded AI agents and third-party applications before committing to new implementations. The priority is to avoid duplicating capabilities, adding unnecessary integration layers or creating competing interfaces for the same business processes.
Build the business case for Snowflake around ERP data requirements. Snowflake on OCI could give Oracle customers another way to analyze financial, operational and transactional data without moving it across cloud environments. ERP leaders should identify which reporting, forecasting or AI use cases require the integration and whether existing tools already meet those needs. Evaluate data consistency, integration complexity, query performance and total cost before changing the architecture.
