Despite surging AI investments, the panel’s most advanced AI employee-facing experiences stop well short of completing end-to-end workflows autonomously.
At most organizations, AI still sits somewhere between answering queries and initiating transactions for simple requests, such as resetting a password or submitting expenses. Last-mile activities rely on employees to stitch together steps manually.
The same disparity shows up in the employee journey. Individual tasks are increasingly automated inside HR, but few organizations can orchestrate hire-to-retire journeys (such as onboarding, promotions, leave, offboarding) from a single request across HR, IT, finance, and other functions. Cross-functional handoffs and disconnected systems are the most common places journeys break.
Underneath both employee experiences and journeys are signs of uneven AI governance maturity and the required mechanisms for autonomous execution. Few leaders say their organizations have common cross-functional alignment between HR and IT in areas such as explicit AI decision rights, shared views of performance and value, and formal ownership of cross-functional workflows.
Design the employee journey beyond automation. Move from AI that assists to AI that completes work end to end through a single AI-powered front door, or conversational AI layer, that turns employees’ natural-language requests into completed work across departments such as IT, HR, finance, and procurement.
Start with employee intent, and map the full path to completion across systems, functions, approvals, and decision points. The goal is not more AI touch points, but fewer handoffs between an employee’s request and the finished outcome.
Then turn governance into executable operating rules by treating IT and HR alignment as a prerequisite: Define who owns the journey, what an AI agent can decide, which actions require human judgment, and which systems must respond automatically. Cross-functional execution won’t work until those rules exist.
The panel says AI-created capacity is already flowing toward strategic or higher-value work and that AI is giving employees more time for meaningful work. But the operating disciplines underneath those claims are much less mature.
For most organizations, capacity gains from AI are either not systematically measured or left to individual managers to decide where they’re directed. In HR, the impact of AI on employee experience is often defined by more visible metrics such as time saved, adoption rates, and cost. Very few say they redirect capacity toward enterprise growth priorities.
When it comes to strategically identifying where AI can transform work, most leaders are focusing on the potential for task automation or augmentation. Far fewer maintain a workforce view linking AI to role redesign, skills, redeployment, or business outcomes.
Start with business outcomes in mind. Work backward from the goals you want to achieve, and determine the tasks and workflows that will get you there. Then identify where AI can augment or automate activity, which work will expand, which roles will change, and which skills will become more important.
As work changes, deliberately reallocate capacity. Start by making your AI deployments visible in a single view of where AI is deployed, how it’s being used, and how much capacity it’s released. Document how freed time will be proactively designed into role and workforce planning rather than quietly reabsorbed.
Most leaders agree that AI adoption works best when managers actively enable it. Manager hesitation or lack of readiness can stall adoption within teams.
However, manager preparation hasn’t caught up with that responsibility. Most organizations are still communicating the AI strategy to managers or training them to use available tools. Far fewer are preparing managers to redesign work, establish new team norms, or formally assess whether managers are ready to lead AI-enabled teams.
This creates a different adoption challenge than a simple skills gap. Enterprises increasingly expect managers to translate AI strategy into everyday work but are still enabling them primarily as technology users rather than as designers of human-and-AI work.
Redesign the manager role for AI. Managers often determine when AI is used, how exceptions are escalated, and how saved time is reallocated. If responsibilities for AI adoption are assigned to managers, those responsibilities should become explicit leadership expectations instead of additional work layered onto an unaltered job.
Enablement should move beyond tool literacy to workflow redesign, capacity allocation, coaching, and employee trust.
Managers also need time and authority to perform that role. Provide a single workspace that takes manual coordination off their plates and automatically displays team signals, open work, and stuck approvals in one place. Enable managers to orchestrate end-to-end journeys for responsibilities they own, such as promotions, onboarding, and transfers, from a single trigger that coordinates cross-system handoffs.
We asked the panel to describe how AI agents are currently planned and managed in their organizations. Most treat agents as software tools assigned to individual tasks or as automated workflow components owned by a single function. A minority plans them as managed digital capacity with defined owners.
Fewer still include AI agents as part of role and workforce planning or as formally designed human-and-AI teams with clear work allocation, governance, and performance accountability.
In addition, the workforce management disciplines for secure and effective AI agent deployment are largely missing. Less than one-third of the panel assign a named business owner, implement training and improvement processes, set performance measures, or define clear escalation and human-override rules. And almost none have processes to formally retire or offboard AI agents.
Govern AI agents as deliberately as people. Decide the human-to-agent mix role by role, assign humans to manage and evaluate performance, and help workers build the systems thinking skills to direct a growing AI agent population.
Govern every agent through a single control layer, including what it can access, what it’s authorized to do, and how it performs. Gain real-time oversight, audit trails, and spend visibility so you can scale your agentic workforce securely and efficiently.
The panel sees a clear employee benefit from AI in work itself. The administrative burden is falling, employees have more time for meaningful and higher-value work, and autonomy is improving for many.
However, those gains don’t translate clearly into the broader employee relationship with the organization. There’s a mixed view of human connection, trust, and fairness. Much of the panel is less confident in future career opportunities for their teams. They also note anxiety regarding monitoring and surveillance.
Yet these concerns are barely addressed by the AI design process. Employee experience receives some consideration before use cases scale, but well-being, autonomy, and trust are often absent as formal scaling criteria. They’re also among the least-used measures of AI value.
The enterprise is therefore measuring whether AI makes work faster long before it measures whether AI-enabled work remains sustainable and trusted.
Make well-being a design requirement. AI use case reviews should assess workload intensity, employee agency, transparency, and perceived career impact before deployment.
Employee trust also needs operational mechanisms: clear human escalation paths, transparency into how AI is used, explicit boundaries around monitoring, and a credible explanation of how roles will evolve as routine work moves to AI.
Our panel of executives knows their organizations are only scratching the surface of the value that AI could bring to HR operations and employee experience.
The enterprises that take a more strategic view of AI adoption, redesigning work to enable more complete interactions and effectively drive employees toward higher-value activity, will be the first to successfully scale and achieve true AI advantage.
The Exec Signals panel comprises 100 technology and business executives from large, global companies, all with decision-making or significant influence over AI and enterprise technology strategy. Insights were gathered through a structured survey, which surfaced directional intelligence on how leaders are navigating AI. Research was commissioned by ServiceNow and conducted independently by global research firm Phronesis Partners.
