Oracle Corp. is expanding the range of generative artificial intelligence capabilities in its Fusion Cloud applications suite to cover uses in enterprise resource planning, human capital management, supply chain and customer experience.

Fusion Data Intelligence, a cross-application data warehouse, is also getting additional generative AI support.

The new features are embedded in existing business workflows and can be extended using the extensibility framework in Guided Journeys, a type of user-configurable copilot. Customers have their choice of large language models.

Oracle now supports more than 50 generative AI use cases within Fusion applications. Security provisions ensure no customer data is shared with large language models or seen by other customers. Access to models trained on proprietary data is limited to individuals. Role-based security is embedded into workflows that only recommend content users are authorized to view.

In the Fusion Cloud ERP suite, new generative features help identify anomalies, variances and biases based on pattern recognition and provide natural-language explanations. Finance professionals can get contextual commentary to explain forecasts produced by predictive models, and program managers can generate executive summaries using details drawn from projects and sub-programs. Generative AI also helps project managers generate tailored plans based on opportunity details, similar past projects and best practices.

Automated product descriptions

In the Fusion Cloud Supply Chain & Manufacturing applications, product specialists can now generate standardized product descriptions with targeted keywords for search engine optimization. Procurement professionals can use generative capabilities to add suppliers to their organization’s supply chain and tap into AI-powered supplier recommendations to reduce costs and risk. They can also generate a customized cover page summary for a specific negotiation.

Users of the Fusion Cloud Human Capital Management application can now use generative AI to build landing pages for different job categories and tailor messaging to specific audiences. AI-powered feedback in Oracle Recruiting gives candidates a better understanding of their job qualifications. A conversational interface helps candidates find answers to common questions about the company, benefits and job-specific requirements. Managers can use generative AI to create surveys.

In the Fusion Cloud Customer Experience applications, contact center agents can now generate summaries of customer chat sessions based on automated transcriptions. Managers can more accurately track ongoing issues and reduce manual tasks for service agents. Salespeople can auto-generate customer success stories from closed-won accounts and produce sales support documentation for sales calls. And marketers can tap into subject line recommendations, email and landing page copy and design recommendations.

AI analytics boost

New features in Fusion Data Intelligence span ERP, SCM, HCM and CX analytics. They give financial professionals better insights into metrics such as collection history, spending, and expense anomalies. One new model flags customers whose payment history suggests likely late payments. SCM models predict the actions needed to ensure on-time delivery, reduce supplier risks, optimize inventory management, and improve warehouse efficiency.

HCM customers can use machine learning to predict staffing needs better, reduce skills gaps, analyze workforce diversity and improve time-to-hire. One model identifies hidden biases in hiring, internal mobility and compensation practices due to gender, age and ethnicity traits. Customer experience professionals can quickly surface relevant offers that boost follow-on sales and reduce customer churn. They can also more accurately forecast sales, optimize pricing, predict churn rates and get data-driven recommendations for product pricing and sales strategies.

A new Fusion Data Intelligence integration with the Fusion Accounting Hub component of the ERP suite allows financial data from multiple accounting systems to be combined to identify correlations and irregularities across balances, journals and sub-ledger transaction details.

Photo: SiliconANGLE