What enterprises should really watch at HP this year

The tech world was taken aback this week by the unexpected departure of HP CEO Enrique Lores, who left the device giant to head PayPal.

But, according to industry experts, the company will fare well against this kind of shakeup as long as it is able to execute on AI PCs, show definitive ROI, differentiate with emerging edge use cases, and strike a balance between pricing and technology advances.

Here’s a look at what enterprises might expect from HP in the coming year.

ROI with AI PCs is still a ‘landmine’

There’s been a lot of hype around AI PCs, and new technologies are arriving regularly, but enterprise adoption remained slow in 2025 due to questions around cost and ROI. HP has pushed hard into the market with its Copilot+ laptops, OmniBook X, EliteBook X, and OmniBook Ultra, powered by Qualcomm Snapdragon X Elite and Intel Core Ultra processors.

Notably, the company recently showed off its EliteBook X G2 Series, with an expanded set of processor options, and has bet on keyboard-based PCs for hybrid workforces.

Still, it’s very early in the game, and enterprises are asking why they should spend extra money to upgrade to AI PCs. Indeed, there are “frilly edge side use cases,” noted John Annand, digital infrastructure practice lead at Info-Tech Research Group, but many IT leaders don’t know where AI fits in their business. Quantifying the actual amount of return remains “nebulous.”

“I just don’t think we’ve found that killer use case for an AI chip on your local PC,” said Annand.

Sanchit Vir Gogia of Greyhound Research partly agreed, noting, “ROI ambiguity is a landmine.” His firm found that, while 57% of CIOs are evaluating AI PCs in refresh cycles, only 19% have approved broad deployment, and even fewer can clearly tie device AI to business KPIs.

This means HP must articulate AI ROI through managed pilots, quantified outcomes, and role-based value, Gogia says. “CIOs should challenge HP and others to show the before-after story,” on factors such as battery life, level of productivity lift, and IT support tickets.

“The new CEO comes in during a very tough time in the PC industry,” noted Anshel Sag, principal analyst at Moor Insights & Strategy. 2026 was poised to be a “huge refresh cycle” with Windows 10 end-of-life (EoL), but due to shortages in memory and other resources, demand is unlikely to be filled due to Microsoft’s “poor timing and messaging.”

No doubt there is a constant push to prove the value of AI and support better on-device experiences, Sag noted. “The good thing for HP is that it has a very diverse ecosystem of suppliers and should weather the storm well.”

Challenges with memory, supply chains

Another big challenge is the memory shortage; Gartner estimates price hikes ranging from 15% to 40% for end users. This is “causing widespread panic purchasing” on one hand, and the decision to “sweat all assets into the grave,” on the other, noted Gartner analyst Autumn Stanish.

“We’re seeing more and more companies decide to just redeploy or purchase non-AI PC refurbished devices to get them through the worst of the shortage, and evaluate the necessity of such an upgrade later into 2027,” she said. HP could take advantage of this trend with its Renew Services portfolio, selling secondary devices at a lower cost, but that doesn’t seem to be a marketing priority.

Enterprises should also monitor supply chain resilience and diversification, Stanish said. She called this latest shortage the “second major supply crisis in the past five years.”

“Keep an eye on communications or announcements from HP regarding what their long-term strategy will be moving forward, as geopolitics, climate, and health crises seem to be increasing in number,” said Stanish.

How HP is differentiating itself in the AI PC market

All this aside, HP is well-positioned once AI PC demand increases.

The company’s strategy stands apart because it focuses on the entire enterprise stack, rather than just chip specs, noted Greyhound’s Gogia. It is bundling AI PCs with telemetry, fleet observability, and security architecture, as evidenced by its Workforce Experience Platform (WXP) and its support for AMD, Intel, and Qualcomm offerings. HP has also been willing to create “unconventional” endpoints like the EliteBoard G1a keyboard PC.

The learning curve matters too, and HP is supporting tuning and governance, “not just shipping hardware with AI stickers,” said Gogia.

His firm predicts that 2026 will be the year HP operationalizes AI PCs as governed, role-specific, edge-capable tools, “not one-size-fits-all miracle machines.” They will also push into tactical edge use cases in remote fieldwork, branch locations, regulated zones, or hybrid roles where latency and connectivity are unpredictable.

Additionally, rather than promoting everything on-device or offloading to cloud, HP is supporting what Gogia described as “a choreography of compute,” in which local NPUs handle lightweight, real-time tasks, while heavier queries go to cloud-based copilots. This means CIOs get better cloud cost control and stronger data compliance.

Further, HP will advance silicon diversity, Gogia noted. Its upcoming Snapdragon-based PCs will be pitched as “ultra-mobile options” with “full security parity” to x86.

Beyond this, Stanish observed, HP’s WXP and its Wolf security chip are positioning HP as a “more holistic” digital workplace provider, rather than just a hardware original equipment manufacturer (OEM).

More broadly across the market, there’s been greater consistency in hardware specifications, and Gartner expects that mainstream models from all vendors will transition from NPUs with lower TOPS ratings to second-generation NPUs with 45-plus TOPS that meet Copilot+ certification standards. As a result, nearly all new devices will be classified as AI PCs, with most achieving Copilot+ certification, and AI enablement on PCs will increasingly extend beyond NPUs, incorporating advanced embedded GPUs from Intel, AMD, and Qualcomm.

HP will likely lean into the practical aspects of AI PCs — better security, enhanced collaboration, self-healing, and general autonomic operations — rather than the more hyped theoretical uses.

Over time, more capabilities will emerge around small language models (SLMs), domain specific language models (DSLMs), light language models (LMs) and other advanced capabilities. However, “we aren’t even close to mainstream for a thing like that yet,” said Stanish.

HP still figuring out how AI can benefit the print category

HP is an undisputed leader in the printer space, which, like other areas of business, is being impacted by AI, albeit not as dramatically.

“AI is relevant for everybody,” said Keith Kmetz, program VP for imaging, printing, and document solutions at IDC. “What the print market is trying to figure out is how AI is going to be applied in products and services in a manner that is differentiated from the conventional device.”

There’s potential for HP and other providers to apply AI to the “aftermarket experience service” and ongoing maintenance, as well as to support cost and general labor efficiencies, he noted.

Memory costs and constraints are hitting all sectors, printers included, and HP and others will likely pass that on to end users, Kmetz predicted. “HP hasn’t been shy in the past of saying ‘Look, we’re going to raise our prices,’” he said. While there aren’t likely to be huge increases, they will “nudge up a bit higher” as component and memory costs rise.

“Anything that needs memory is going to be faced with this increased cost that, frankly, most are going to expect,” said Kmetz.

Strategy is intact

The HP CEO transition is not the risk factor most CIOs think it is, Gogia noted. “The board has signaled continuity,” and “the strategy is intact.” However, CIOs should keep an eye on HP’s execution cadence, integration roadmap, and field responsiveness.

To see success with AI PCs, HP must continue to demonstrate measurable value — ticket reductions, productivity gains, and strong device observability — while balancing innovation with disciplined pricing, said Gogia. AI premiums are creeping into “endpoint TCO,” and the company’s pricing strategy will indicate whether it is prioritizing profits or defending its margins.

Furthermore, enterprise leaders should watch whether HP maintains a multi-silicon strategy. Supporting Qualcomm, AMD, and Intel is “smart,” Gogia noted, but only if manageability, patching, and telemetry remain consistent. Fragmented models break user trust.

Customers should ask HP how much value is delivered rather than just promised, and monitor whether support is consistent across regions and if partner responsiveness dips. This is also a good time to renegotiate service level agreements (SLAs), align roadmaps, and validate tech stacks, Gogia advised.

“CIOs should treat HP’s new leadership year as a stress test,” he said.

This article originally appeared on Computerworld.