That transition was highlighted by Hiten Bhuta during SHRM Atlanta's HR Forward: AI & Workforce Innovation Summit, held September 18 at Georgia Tech's CODA Tech Square in Atlanta.

During the event's Lunch & Tech Demos session, Bhuta demonstrated practical applications of AI agents for employee onboarding, HR policy questions, and workforce analytics.

The demonstrations addressed a broader question facing HR and technology leaders: If AI agents can complete tasks across multiple enterprise systems, will employees still need to interact directly with every underlying application?

“The seat is no longer the only unit of value. AI agents are shifting HR software toward workflows and outcomes, while payroll, compliance and human judgment remain essential,” said Hiten Bhuta.

Enterprise HR Software Is Entering a New Phase

The scale of today's HR technology market makes this shift significant.

ADP reports more than 1.1 million clients across more than 140 countries, with payroll processing covering more than 42 million workers worldwide. The company reported $21.9 billion in fiscal 2026 revenue.

Workday, another major provider of enterprise HR and financial software, reported $9.552 billion in fiscal 2026 revenue and said its platform processed 1.7 billion AI actions during the fiscal year.

During Workday's third-quarter fiscal 2026 earnings discussion, company executives also said that 75% of new sales included an AI solution.

These developments point toward a broader change in how enterprise software may be consumed.

From Software Seats to Completed Work

For decades, enterprise software economics have largely been based on the number of employees authorized to use a system.

AI agents introduce another possibility.

An agent can interact with software through APIs and automated workflows, potentially completing tasks without requiring an employee to manually navigate every application.

In July 2026, Gartner estimated that up to $234 billion in enterprise application SaaS spending could be exposed to what it calls “agentic arbitrage” by 2030. The estimate applies to enterprise applications broadly rather than HR software specifically, but the underlying concept is relevant to workforce technology.

As agents take on more tasks, companies may increasingly evaluate software based on work completed, usage, and business outcomes, alongside traditional user-based licensing.

Several technology companies are already experimenting with alternative pricing structures. Workday has described a hybrid consumption model for AI through its Flex Credits approach, while Zendesk has introduced outcome-based pricing for AI-agent resolutions. GitHub Copilot also moved toward usage-based AI credits in 2026.

What AI Agents Could Handle in HR

The Atlanta demonstration focused on practical workflows rather than automation for its own sake.

Several HR processes could potentially benefit from AI-assisted workflows, including:

  • Employee onboarding questions

  • HR policy retrieval

  • Information routing

  • Recurring administrative requests

  • Workforce data analysis

  • Employee self-service interactions

An AI agent could potentially retrieve information from existing systems, guide employees through routine processes, and route more complex requests to the appropriate HR team.

However, not every HR function is equally suited to autonomous automation.

Payroll, regulatory compliance, sensitive employee decisions, confidential records, and systems of record require strong controls around accuracy, privacy, auditability, governance, and human oversight.

AI Adoption Is Also Changing Technology Companies

The transition toward AI-driven enterprise software is affecting technology vendors themselves.

Workday announced a restructuring in February 2025 that initially cont