PwC: 5 Actions CHROs Must Take for Agentic AI in HR

The CHRO's Agentic AI Moment Has Arrived

A new PwC report reveals that 82 percent of HR leaders plan to deploy agentic AI in at least one HR function by mid-2026. The survey, covering 1,200 CHROs and senior HR executives across 40 countries, signals that human resources is no longer a lagging adopter of enterprise AI. It is becoming one of the most active deployment targets.

The shift is driven by unmistakable pressure. Talent acquisition costs have risen 35 percent since 2023. Employee turnover in knowledge work remains elevated. HR teams are expected to do more with flat or shrinking headcount. Meanwhile, the complexity of compliance, from pay transparency regulations to AI hiring laws, continues to compound.

PwC's report outlines five critical actions that CHROs must take to deploy agentic AI effectively, avoid common pitfalls, and position HR as a strategic driver of organizational transformation rather than a cautious follower.

Action 1: Redesign Workflows Before Deploying Agents

The most common mistake in HR AI adoption is automating existing workflows without rethinking them. PwC found that organizations that deployed AI agents on top of legacy HR processes achieved only 15 to 20 percent of the potential value. Those that redesigned workflows around agentic capabilities captured 60 to 80 percent.

flowchart LR
    INPUT(["User intent"])
    PARSE["Parse plus<br/>classify"]
    PLAN["Plan and tool<br/>selection"]
    AGENT["Agent loop<br/>LLM plus tools"]
    GUARD{"Guardrails<br/>and policy"}
    EXEC["Execute and<br/>verify result"]
    OBS[("Trace and metrics")]
    OUT(["Outcome plus<br/>next action"])
    INPUT --> PARSE --> PLAN --> AGENT --> GUARD
    GUARD -->|Pass| EXEC --> OUT
    GUARD -->|Fail| AGENT
    AGENT --> OBS
    style AGENT fill:#4f46e5,stroke:#4338ca,color:#fff
    style GUARD fill:#f59e0b,stroke:#d97706,color:#1f2937
    style OBS fill:#ede9fe,stroke:#7c3aed,color:#1e1b4b
    style OUT fill:#059669,stroke:#047857,color:#fff

The difference is structural. Legacy recruiting workflows, for example, involve a recruiter manually screening resumes, scheduling phone screens, conducting initial interviews, coordinating with hiring managers, and managing offer logistics. Deploying an AI agent to assist with resume screening within this workflow produces modest gains.

Redesigning the workflow means asking: if an AI agent could autonomously source candidates, assess qualifications against job requirements, conduct structured initial assessments, schedule interviews with hiring managers, and manage offer letter generation, what should the recruiter's role become? The answer is that recruiters shift from administrative processing to relationship building, candidate experience management, and strategic workforce planning.

PwC recommends that CHROs conduct workflow redesign workshops for each HR function before selecting or deploying any AI agent technology. These workshops should involve HR practitioners, hiring managers, IT, legal, and employee representatives to ensure that redesigned workflows serve all stakeholders.

Action 2: Upskill HR Teams for an Agent-Augmented World

PwC's survey found that 67 percent of HR professionals feel unprepared to work alongside AI agents. The skills gap spans three dimensions:

• Technical literacy: HR professionals need to understand what AI agents can and cannot do, how to evaluate agent outputs, and how to configure agent behavior for different scenarios. This does not require coding skills, but it does require comfort with data-driven tools and an understanding of AI capabilities and limitations
• Judgment and oversight: As agents handle routine tasks, HR professionals must develop stronger judgment for the complex, ambiguous situations that agents escalate. This includes bias detection in agent recommendations, ethical assessment of automated decisions, and the interpersonal skills needed for high-stakes conversations that agents cannot handle
• Strategic capabilities: With agents handling operational work, HR teams can invest more time in workforce planning, organizational design, culture development, and change management. These strategic capabilities need to be developed proactively, not discovered retroactively after agents are deployed

PwC recommends that CHROs allocate dedicated upskilling budgets and create structured learning paths that prepare HR teams for agent-augmented roles over 12 to 18 months.

Action 3: Establish Governance Before Scaling

HR AI governance is uniquely complex because HR decisions directly affect people's livelihoods. An AI agent that makes a flawed hiring recommendation, an unfair compensation decision, or an inaccurate performance assessment can cause real harm to individuals and expose the organization to legal liability.

PwC outlines a governance framework with four pillars:

• Transparency: Employees and candidates must know when AI agents are involved in HR decisions that affect them. This is not just ethical best practice. It is increasingly a legal requirement under regulations like the EU AI Act, New York City's Local Law 144, and Illinois's AI Video Interview Act
• Bias auditing: AI agents used in hiring, promotion, and compensation decisions must undergo regular bias audits that measure outcomes across demographic groups. These audits should be conducted by independent parties, not the teams that built or deployed the agents
• Human oversight requirements: The governance framework must specify which decisions require human review before execution. PwC recommends that all termination, compensation, and promotion decisions involving AI agent input include mandatory human review, regardless of the agent's confidence level
• Appeals and redress: Employees and candidates must have clear mechanisms to challenge AI-influenced decisions and receive human review of their cases

Action 4: Measure ROI Rigorously and Honestly

PwC found that only 23 percent of organizations deploying HR AI have established clear ROI measurement frameworks. Without rigorous measurement, organizations cannot distinguish between agents that deliver genuine value and those that create the appearance of efficiency while introducing hidden costs.

Effective ROI measurement for HR AI agents includes:

• Time savings quantification: PwC's data shows that AI agents can reduce recruiter time on sourcing and screening by up to 70 percent. But this metric only matters if the saved time is redirected to higher-value activities. If recruiters spend saved time on other administrative work, the organizational ROI is minimal
• Quality impact measurement: Are candidates hired through agent-assisted processes performing better, ramping faster, and staying longer than those hired through traditional processes? These downstream metrics take 6 to 12 months to materialize but are the true measure of agent value in recruiting
• Employee experience tracking: AI agents that improve HR efficiency but degrade the employee experience, for example through impersonal onboarding interactions or frustrating chatbot experiences, may create long-term retention costs that exceed their short-term savings
• Compliance cost avoidance: Agents that reduce compliance errors, ensure consistent policy application, and maintain proper documentation can avoid significant regulatory penalties and litigation costs

Action 5: Scale Incrementally with Continuous Learning

PwC warns against the "big bang" approach to HR AI deployment. Organizations that attempt to deploy agents across multiple HR functions simultaneously typically experience implementation fatigue, change resistance, and quality problems that undermine confidence in the technology.

The recommended approach is incremental scaling:

• Start with a single, high-impact use case: Most organizations achieve the best initial results with recruiting or employee onboarding, where processes are well-defined and ROI is measurable
• Prove value before expanding: Demonstrate clear, measurable ROI in the initial use case before investing in additional agent deployments. This builds organizational confidence and executive support for broader adoption
• Build internal capability: Each deployment builds skills, governance processes, and technical infrastructure that make subsequent deployments faster and lower risk
• Incorporate feedback loops: Agents should improve continuously based on feedback from HR professionals, hiring managers, employees, and candidates. Organizations that treat agent deployment as a one-time project rather than an ongoing optimization effort see diminishing returns

The Stakes for CHROs

The PwC report concludes with a clear warning: CHROs who wait for agentic AI to become a fully proven, risk-free technology will find themselves managing increasingly uncompetitive HR operations. The organizations that move now, with proper governance, measurement, and change management, will build capabilities that compound over time. Those that wait will face the dual burden of catching up on technology adoption while competing for talent against organizations where AI agents have already transformed the candidate and employee experience.

The 82 percent adoption intention figure suggests that inaction is no longer the default position. The question facing most CHROs is not whether to deploy agentic AI, but whether they will do it well.

Frequently Asked Questions

What HR functions are best suited for initial agentic AI deployment?

PwC recommends starting with recruiting or employee onboarding. Recruiting offers high-volume, repetitive tasks with clear success metrics such as time-to-fill, cost-per-hire, and quality-of-hire. Onboarding involves structured, multi-step processes with multiple system touchpoints that agents can orchestrate efficiently. Both functions provide measurable ROI within three to six months of deployment.

How can CHROs address employee concerns about AI replacing HR jobs?

Transparency and early involvement are critical. CHROs should communicate clearly that AI agents are intended to handle administrative and repetitive work, freeing HR professionals for higher-value strategic tasks. Involving HR team members in workflow redesign workshops gives them agency over how their roles evolve. Dedicated upskilling programs demonstrate organizational investment in the HR team's growth rather than replacement.

What legal risks exist for AI agents in HR decision-making?

AI agents involved in hiring, promotion, compensation, and termination decisions face scrutiny under employment discrimination law, data privacy regulations such as GDPR and CCPA, and emerging AI-specific legislation like the EU AI Act and New York City's Local Law 144. Organizations must conduct bias audits, maintain transparency about AI involvement in decisions, ensure human oversight of high-stakes choices, and provide appeals mechanisms for affected individuals.

How much time can AI agents save in the recruiting process?

PwC's data shows that AI agents can reduce recruiter time on sourcing and screening by up to 70 percent. Interview scheduling automation saves an average of 5 hours per hire. Offer letter generation and onboarding coordination can be reduced from days to hours. However, the total organizational ROI depends on how the saved time is redirected and whether the quality of hiring outcomes improves alongside the efficiency gains.