How can “AI-in-the-loop” relieve overextended IT teams by offloading repetitive, long-tail tasks while keeping human judgment in charge?
To address that, I spoke with Faisal Masud, President, HP Digital Services, who outlined business wins—proactive device remediation, fewer tickets, and lower support costs—paired with a mandate to improve the employee experience.
Masud positions HP’s Workforce Experience Platform (WXP) as an AI-powered, hardware/OS-agnostic way to secure endpoints, manage global fleets in real time, and continuously gauge employee sentiment. Agentic AI adoption, he said, is rising but uneven; enterprises will integrate agentic AI pragmatically, with humans remaining accountable for the foreseeable future.
Core Takeaways
- AI-in-the-loop, not AI-all-the-way: Use AI for the “inform” and “consult” stages while humans stay responsible for approvals and decisions.
- Employee experience is a business KPI: Proactive, pattern-based fixes reduce tickets and onboarding friction so staff can focus on higher-value work.
- Operational and financial gains: Automated fleet health, preemptive remediation, and data-driven device right-sizing cut desk-side and procurement costs.
- WXP in practice: WXP is a HP’s multi-tenant, cloud-native platform that embeds endpoint security, real-time fleet management, and employee feedback—deployed in weeks and agnostic to hardware/OS; built for global enterprises and channel-led SMBs.
Key Quotes
Automation that gives humans their time back
“There’s many ways to look at this, but I would say any of those long tail tasks that most humans don’t want to do and they can be handled through automation, the vast majority of people will say they would be delighted to take that off their plate so they can focus on things that are way more important, more strategic….
“If you look at it from that perspective, I think there’s a lot of goodness there. But of course there’s the other side too…I think the other side is Microsoft put out this list yesterday on X on the most threatened roles in the world…customer service goes to the top of the list, and very understandable… But there’s the other side which is more like, well, is this threatening the livelihood of folks as it emerges as the smartest way to do everything? And I think we’re pretty far from that, but right now it’s super helpful to have people have a lot more bandwidth to do things that are important.”
Where AI fits in the RACI framework
“I wrote about this recently—this concept of AI in the loop—and the framework of RACI (Responsible, Accountable, Consulted, and Informed)… Assuming that AI is going to take over the entire process from thinking of a problem all the way down to resolving the problem without having a human in the loop feels like a bridge too far…
“It’s almost the reverse: in the RACI framework you could have the ‘inform’ and ‘consult’ be the happy place for AI, where you don’t really need to have humans do that work that could be generated through deep research, which you could find on any LLM. I feel like it’s a good balance of having humans involved while really leveraging AI in the process.”
Proactive fixes that lift employee experience
“We can boost employee experience—for example, a traditional employee joins on day one and in the first week they’re filing tickets: What’s this? What’s that? AI is really good at pattern recognition. It’s seen these problems before; the data has been trained to know that if I spot this, then these are the different permutations to solve.
“The velocity and quality gained by using AI to resolve those things is why you rely on it… I’m not having to file a ticket because AI just popped up on my screen and said, ‘Hey, there was a problem coming up. I took care of it—just FYI,’ versus going through ticketing or desk-side because my browser’s got a lot of latency—why? And if you’re a CIO procuring for a hundred-thousand-device fleet, AI can do a lot of that based on user history, actual device usage, and cost. You gain time and save money—and in the end, the output is experience.”
Why human judgment stays in the loop
“Agentic AI is getting a lot of [focus], but to think this becomes a full handoff is a stretch. We saw this with Klarna—they moved all their customer service to AI and then moved it back to humans… Back to the RACI concept: AI serves a lot of purposes, but it doesn’t serve every purpose. Human judgment is still needed in the loop.
“If you don’t have skin in the game, you’re not invested in the decision, and AI has no skin in the game—what’s AI’s skin in the game? None. We have skin in the game. So the human in the loop with the AI is where this goes. I don’t see this becoming a 100% handoff in three years—especially at the enterprise level.”
