Wipro’s HR AI agent cut responses from 48 hours to five seconds

Wipro says its custom agentic AI assistant has collapsed average response time to employee queries from 48 hours to five seconds. The assistant was co-created with enterprise agentic AI platform Ema Unlimited, and Wipro designed it to assume responsibility for 50 HR tasks that previously fell to human employees. This is not “AI as a search box.” Jayaswal frames it as AI agents that can navigate complex internal systems and take action in the flow of work.

That shift is exactly why leadership teams are treating agentic AI as a workforce issue, not just a productivity lever. The MIT Technology Review piece points to projections that adoption of AI agents could surge by as much as 300% over the next two years, and early deployments in customer service, HR, and sales have already delivered productivity gains of 30-50%. In other words, this is moving from pilots to operations quickly, and HR is on the front line because the work agents take over touches roles, identity, and culture.

Jayaswal’s key argument is that AI agents force a role redesign, not a thin layer of automation. He estimates that three-quarters of current roles will require redesign, reskilling, or redeployment by 2030 due to agentic AI. For leadership, the point is not simply to reduce headcount or “offload tasks.” It is to redeploy employees toward higher-value work that combines creativity and cross-functional collaboration. In Wipro’s case, the agent handles “rote administrative tasks” such as sorting timesheets and helping employees navigate policies, then enabling actions. Humans get more time for work that is harder to mechanize and harder to replicate: interpreting context, coordinating across teams, and making judgment calls.

But the handoff from human to agent is where many enterprises could stumble, and Wipro’s caution is explicit: when an AI agent touches organizational and personal data, it needs stricter guardrails than consumer AI. Jayaswal emphasizes that once you expose an agent to organizational data and integrate it into multiple enterprise systems, the “pathways around the AI agent become extremely important.” Put plainly, it is not enough to make the agent smart. You also have to constrain what it can do, where it can go, and how it can behave across systems that contain sensitive information.

That is why governance becomes a management muscle, not an IT checkbox. Jayaswal suggests building governance layers such as an AI council. He also flags that organizations have to treat accountability and “pathways” as first-class design problems. This connects to a broader trust issue noted in the same MIT Technology Review content: many organizations define AI agents as teammates or colleagues on org charts, but new research says this could erode trust and professional identity. When employees view an agent as a peer, it can blur ownership and responsibility, raising uncomfortable questions about who is accountable for decisions and outcomes.

On the skills front, leadership should expect hiring and training to change shape. More than four in five HR leaders say they’re planning to reskill workers to stay competitive in an AI-agent-shaped market. Technical skills matter more, and leading employers including Salesforce, Danone, and Walmart are already rolling out dedicated AI and digital skills programs for everyone from frontline workers to C-suite executives to build baseline AI literacy. At the same time, Jayaswal stresses that desirable soft skills will evolve in a very practical way: employees who assign tasks to an AI agent need to articulate modular steps, specify desired outcomes, and set parameters or guardrails so the agent does not access or share confidential data.

Culture and well-being are the third rail. The article notes confusion and knowledge gaps, with 73% of HR leaders reporting that employees don’t yet understand how digital labor will impact their work. And even if productivity rises, there can be a human-loss effect: more interactions with AI agents can reduce the “human touch” previously provided by service delivery partners, leaders, or peers. Jayaswal argues that employee services should encourage social connection and empathetic communication to help teams navigate blended work. He also points out that HR leaders see this as a changing-of-the-guard moment: more than three-quarters of HR leaders believe AI agents will transform workplace norms, requiring a complete reappraisal of how roles and responsibilities are distributed, how skills are prioritized, and how culture is shaped. And 86% of chief HR officers predict that navigating digital labor shaped by agentic AI will be central to their role.

So what does this mean for the rest of the leadership table, beyond HR? The content lands on a simple but intense thesis: agentic AI is poised to scale at breakneck speed, and leadership teams now have to decide how to adapt their organizational strategies to optimize both tech gains and the employee experience. If you are a CEO, board member, CFO, or COO, the second-order risk is that you get the efficiency but lose the trust, the governance discipline, or the people systems that make adoption stick. Wipro’s five-second turnaround is the headline. The real takeaway is that getting from “agent as tool” to “agent as collaborator” requires rethinking roles, governance, manager skills, and culture, all at once.