SAP Rolls Out Joule Work and Expands AI Agents Across Core Operations

Enterprise software provider SAP has started the commercial rollout of Joule Work, shifting its AI assistant from a conversational sidekick into an active operational workspace. Announced at the SAP Connect conference in Las Vegas, the release gives corporate users an interface where software agents can carry out cross-departmental tasks directly inside core business systems.

Alongside the interface rollout, SAP introduced SAP Pay, an integrated payment tool designed to settle and reconcile invoices automatically within cloud enterprise resource planning workflows. The software giant also unveiled fresh domain-specific Joule Assistants for finance, human resources, and supply chains, alongside expanded integrations to track supplier financial risk.

Moving from chat prompts to autonomous tasks

Rather than asking staff to query a standard chatbot window, Joule Work acts as a coordinating layer. It connects to the company's underlying knowledge graph—which maps millions of enterprise data fields—to give automated agents the context required to carry out multi-step jobs.

Before opening access to external clients, SAP tested the environment across 110,000 internal staff. The vendor reported internal productivity improvements of around 20 percent across routine finance, procurement, and administrative operations. Early corporate testers reported that multi-step tasks across their enterprise resource planning environments, which previously took up to an hour of manual navigation, could be completed in minutes.

To help organisations manage the growing collection of automated workers, SAP is also making agent monitoring and governance tools available through its administrative portal, allowing companies to control how much independence each agent has.

What this means for Australian businesses

For Australian companies, particularly mid-market firms and growing enterprises running standard business platforms, this release signals a major shift in how workplace AI will arrive:

• Automation is embedding into existing systems: You do not need to build custom software to take advantage of agentic AI. Mainstream business systems like enterprise resource planning and accounting platforms are building autonomous workflows directly into the software your staff already use every day.
• Process redesign matters more than prompt engineering: As software takes on multi-step administrative tasks like invoice reconciliation, vendor onboarding, and leave processing, business owners should review internal approval chains. Ensure clear financial limits and oversight checkpoints are in place before handing over transactional autonomy.
• Look for measurable time savings in back-office work: Small and medium enterprises spend substantial staff hours shuffling information between spreadsheets, emails, and inventory platforms. Keeping an eye on how your current software vendors are deploying autonomous agents can help you identify manual data-entry bottlenecks to eliminate over the coming financial year.