Share with your CHRO
Workday is betting that HR and finance software wins by disappearing into the tools employees already use, not by demanding they log in somewhere new. The company’s expanded partnership with Google Cloud puts Workday’s Sana Self-Service Agent inside Gemini Enterprise, letting employees check leave balances, approve timesheets, and query expense policies through a conversational interface without leaving Google’s workspace. Gemini becomes the default model powering Sana. Alphabet itself is an early internal user, building a custom Workday agent for its own administrators.
What this means for your business
If your organization runs Workday for HR and Google Workspace for daily work, this partnership compresses a decision you were already going to face. The question isn’t whether to connect these systems eventually; Workday is now treating that connection as table stakes and moving the timeline up. CHROs whose employees still toggle between a Workday portal and a Google inbox to answer basic HR questions are looking at a near-term window to eliminate that friction, and the organizations best positioned are those that have already standardized on both platforms.
The deeper architectural move here is what Workday calls agent-to-agent workflows, where an AI agent hands off a partially completed task to another agent without a human restarting the process from scratch. Think of it as a relay race for software, where each leg passes a baton rather than making the runner retrace the whole course. For HR leaders, this matters because it’s the mechanism that lets an employee’s leave request in Gemini trigger a payroll input update in Workday without a manager touching a form. The zero-copy data integration with Google Cloud Lakehouse, where Workday data stays in place rather than being copied into a new environment, adds a governance layer that should reduce the compliance friction that typically kills these cross-system projects before they scale.
Workday making Gemini the default model for Sana is a meaningful commitment, not a neutral technical choice. It ties Workday’s AI quality floor to Google’s model roadmap, and while Workday says customers can swap the model, the default is where most deployments will stay. CHROs negotiating enterprise renewals with Workday in the next 12 months should treat the Gemini integration tier as a line item worth scrutinizing, because the value of the self-service agent is directly proportional to the model underneath it. If Google’s models slip relative to competitors in reasoning quality, Workday’s HR automation story slips with them.
Concept deep-dive: Model Context Protocol
Model Context Protocol, or MCP, is a technical standard that lets AI agents share the context of a conversation or workflow with other agents or applications, the way an email thread keeps everyone in a room on the same page. Without it, each agent starts from scratch. With it, an agent handling a payroll question can hand off mid-task to one that manages approvals, passing along what it already knows. In HR workflows, MCP is what makes multi-step processes feel like one continuous interaction rather than a chain of disconnected prompts.
Based on reporting from Workday & Google Cloud deepen AI agent partnership, originally published 2026-06-01 03:00:00.
