HPE customers on agentic AI: No, you go first

But like cloud computing and digital transformation, this may be a buzzword they can't ignore forever

HPE Discover 2025 HPE envisions a future where customer systems are filled with its agentic AI products, but reactions from the HPE Discover show floor in Las Vegas this week suggest the company has a way to go to convince folks to buy in.

We spoke to a number of HPE customers following CEO Antonio Neri's agentic AI-focused keynote, and while we found plenty who were enthusiastic about the potential future of agentic AI, their enthusiasm was tempered by a nearly universal desire not to be the first ones to try it. 

For those not familiar with agentic AI, it's currently the official Next Big Thing in the AI revolution. Rather than just being able to gather data and respond like your typical LLM chatbot, AI agents are basically souped-up workflow bots. They can autonomously perform most actions where a root cause analysis requires a specific action, for example, with humans in the loop now optional.

As one healthcare tech professional told us, agentic AI shows promise, but not enough to warrant experimenting with it now, especially not in the healthcare industry, with its many data privacy rules hovering over the shoulders of technology teams. 

"Many businesses have the availability to begin adoption and testing," Joseph Gravish, an IT professional with life sciences firm Eversana, told us from the show floor. "For other use cases like healthcare, it's not complete or finished in a wrapped-up way businesses can adopt." 

Privacy concerns were echoed by others we spoke to from various industries - even tech professionals in the marketing space told us they were worried that agentic AI, autonomous or not, wasn't going to be a good fit for their organizations. Part of that, HPE customers shared, had to do with a lack of explanation of data privacy rules - they might not be dealing with the Health Insurance Portability and Accountability Act (HIPAA), but customers still have an expectation that proprietary data isn't being fed into a third-party AI. 

Many of us are sitting on the sidelines and waiting to see full functionality

We also asked attendees whether they'd heard enough from their fellow tech professionals to get a sense of a broader opinion on the agentic AI announcements. As with what we heard directly, most said the scuttlebutt was generally optimistic, but still with a consistent wait-and-see attitude. 

"I think the general sentiment from everybody is that it's going in the right direction," Gravish told us. "But many of us are sitting on the sidelines and waiting to see full functionality." 

While some aspects of HPE's agentic AI vision - like plugging it into the company's entire software stack - are still coming soon, that's not the case for its Aruba networking agent automation, executives claimed during the event. Unsurprisingly, of the people we spoke to at the show, network engineers were the most gung-ho about HPE's agentic AI announcements. 

One network engineer, who works in the education sector in the UK and asked not to be named in this article, told us that the Aruba announcements came at the perfect time - his organization is in the midst of a tech refresh, and those AI agents sound lovely, thank you very much. His team plans to make extensive use of Aruba Central AI agents, and he expressed a belief that the technology is mature and ready for network operations. Outside the networking space, he added, it's a different animal entirely.

In other words, HPE can be happy that its announcements pleased Aruba customers, but it might be pushing too hard, too fast for other buyers. 

"The market is hungry for agentic transformation," senior AI practice director Mike Trojecki from infrastructure firm World Wide Technology said during the HPE Discover partner growth summit. But Trojecki also admitted that HPE is still working to develop next-generation AI use cases for its customers.

We only get what they give

Speaking to reporters at a press briefing after the keynote, Neri said that HPE customers are not embracing the potential benefits of AI because they're unwilling to fail.

That's easy for Neri to say when he's not the one whose company is on the line by deploying an AI that could accidentally expose sensitive customer data, or cause more headaches than it resolved, but some others agree. 

The question customers should be asking is how quickly and expertly can they embed AI and agentic AI into their business processes to improve overall customer experience before their competitors overshadow them

Cyndi Privett, owner and principal analyst at channel analytics firm Viewpoint Research, told us in a conversation at Discover that those waiting to see results come in risk of being left behind. 

"The question customers should be asking is how quickly and expertly can they embed AI and agentic AI into their business processes to improve overall customer experience before their competitors overshadow them," Privett told us. 

If that sounds familiar, you've probably been around long enough to have seen the beginning of the cloud era, or maybe the earliest stages of the digital transformation years. Feel free to insert your own adapt-or-die moments from the past few decades. 

Doubting the need to be an early adopter is fair, but we only need to look back to the COVID-19 pandemic to see how different tech adoption philosophies shook out, Privett explained. 

"In March 2020, businesses around the world had to pivot to a more digital presence almost overnight and those that had invested early in digital storefronts and ecommerce were ready," Privett said. As was the case five years ago, early movers are going to have the advantage, regardless of whether we face another sudden need to adapt. 

"Partners and customers that are skilled and invested in AI practices and deploying agents will be positioned to harness the opportunity 2025 and 2026 presents," Privett predicted. "If they don't they will be left behind quickly."

That may not be enough to convince the many HPE customers who were less than impressed at the current state of the company's agentic AI offerings. However, when put in the context of all the other agentic AI buzz in the first half of 2025, skepticism may not matter: Yet another OEM has spoken, and AI agents are going to be everywhere soon enough. 

World-altering event or not, the enterprise space may be rapidly approaching another adapt-or-die moment. ®