The Shift Toward Edge AI To Protect Data Privacy And Personalize The Workplace Experience

Having travelled halfway across the world from Palo Alto, David McQuarrie, HP’s Chief Commercial Officer, sat down with host Hao Tran in this episode of the Vietnam Innovators podcast. Since David’s first visit to Vietnam in the early 2000s, the country has undergone a remarkable transformation. Once an agriculture-driven economy on the cusp of modernization, Vietnam has become increasingly robust both economically and technologically. Today, its public and private sectors are turning their attention to the next technological revolution: AI transformation, and more specifically, the shift from cloud-based computing to local processing.

In this episode, David discusses the latest developments in local AI, HP’s contributions to this technological shift, and how AI adoption can help Vietnam accelerate its economic growth and bridge the gap toward its GDP targets.

Customers Across The Spectrum Are Adopting AI Or Preparing To Do So

AI is no longer a question of “if”; it is now deeply embedded in business conversations. HP serves a diverse range of customers, including public-sector organizations, large enterprises, small businesses, and individual consumers. Customers are either already embracing AI or looking to adopt it, each for their own reasons.

HP’s enterprise customers, in particular, are under increasing pressure to move beyond pilot projects and deliver tangible business results. To make this transformation effective, organizations must look beyond applying AI to individual tasks or isolated steps of a process solely to cut costs. Instead, they need to consider how AI can create value across the business as a whole.

Rather than simply replacing each human step with an AI agent or some form of robotic automation, organizations should be willing to rethink and redesign processes end to end. They should keep sight of the outcomes they want to achieve and the value they aim to deliver to customers.

Another important development is the growing ability to customize AI software for specific industries and local markets. Small and medium-sized enterprises (SMEs), in particular, may be able to adopt and adapt these solutions more quickly because they often have fewer layers of decision-making and more flexibility to experiment.

When Data Security Becomes A Make-or-Break Question

The Vietnamese Law on Personal Data Protection, which takes effect in 2026, has started a lasting conversation about data privacy for both individuals and organizations. Yet, with vast amounts of information stored in the cloud, the conversation around data security remains largely focused on preventing potential leaks. HP’s current priority is not simply to strengthen defenses against data breaches, but also to ensure that sensitive data can be processed and kept on device.

In a sense, the world is witnessing a reversal. Twenty years ago, most information was stored in on-premises facilities. The cloud revolution then moved much of that data and many computing workloads to remote infrastructure. Now, growing concerns about information security are prompting organizations to bring certain workloads back to local devices and on-premises systems.

This change is particularly relevant in the AI era, when organizations need to understand every party in their AI ecosystem - from cloud and hardware providers to model developers - to determine how their data is handled and protected. The cloud remains an important part. However, looking ahead, some AI workloads will be better suited to local processing. This approach keeps sensitive data from being used to train competitors’ models and can offer significant cost benefits in some cases.

The Cost-Benefit Case For Shifting Toward Local AI And NPU-Powered AI PCs

Many organizations are asking whether they should invest in local AI devices while most still rely on the cloud. In David’s view, the choice is not between the cloud and local devices but how to make the most of both. The key question is: which workloads belong where.

The cloud offers scalability, while local AI delivers speed, privacy, and responsiveness. Ultimately, it comes down to understanding the cost-benefit equation and future-proofing the business.

The cost of using cloud-based AI models is becoming a growing consideration. In recent years, the market has seen sustained price increases across electronic devices, driven in part by rising data-center demand, supply-chain pressures, and broader geopolitical factors. As demand for computing power continues to grow, prices are expected to remain under pressure.

As organizations scale their AI usage, they may face higher inference costs and usage-based charges, as well as caps on the number of tokens available under certain pricing plans. The more often an organization uses AI, the faster its inference costs can add up.

This is driving some businesses to consider moving suitable workloads to local AI models, where the upfront investment in hardware can reduce recurring usage costs. The decision is increasingly an economic one, rather than a purely technical one.

The implications vary across industries. Sectors such as healthcare and banking have long operated under stringent data-security requirements. These organizations already possess valuable first-party data and may not need to rely on general-purpose large language models (LLMs) for every application. Instead, they may benefit more from models tailored to their organizational or industry needs, which offer appropriate data protection and can run locally.

The market is moving in this direction regardless. Gartner forecasts that AI PCs will account for around 55% of the worldwide PC market in 2026. At HP, AI PCs represented 46% of the company’s shipment mix in the previous quarter, with the share expected to reach 50% by the end of the fiscal year.

Adopt, Adapt, Or Abandon: Would AI Replace Human Workers Or Augment Their Workflow?

Early in the conversation, much of the discussion focused on whether AI would replace human workers. In reality, certain tasks traditionally performed by people will indeed be automated.

However, a more likely scenario is that AI will not replace human workers wholesale. Instead, workers who know how to leverage AI will replace those who do not. This is no different from the advent of electricity, the Industrial Revolution, or the internet. Each radical technological shift transformed the way people worked.

AI also has the potential to flatten traditional organizational hierarchies. Some of the most enthusiastic adopters of AI are younger workers. In these cases, AI is increasingly taking over tasks that traditionally helped workers build expertise over time. That expertise provides the context needed to assess and apply AI-generated outputs effectively. Without that know-how, workers may struggle to distinguish useful insights from inaccurate or unsuitable results.

What we are discovering, therefore, is that human expertise and contextual understanding remain essential. At least in the coming years, AI is unlikely to replace the accumulated knowledge and judgment that skilled knowledge workers bring to the table. Instead, it can take over much of the routine, lower-value work, allowing these professionals to spend less time on repetitive tasks and apply their expertise where it matters most.

This creates an opportunity to bring two generations of workers together: experienced professionals who possess deep industry knowledge and younger employees who are enthusiastic about AI and eager to explore its applications. When these strengths are combined through mutual mentorship, they can become a powerful force for innovation.

Vietnam’s Future In AI Developmental Landscape

In David’s assessment, the Vietnamese government has embraced the AI transition more readily than some of its Western counterparts. This openness is driven by factors such as demographics, perceptions of risk, national policies, and a greater entrepreneurial willingness to experiment with new technologies.

In a global context, embracing AI could play a significant role in helping Vietnam achieve its development goals. As a powerful equalizer, AI has the potential to help Vietnam narrow the gap with more developed economies, accelerate innovation, and unlock new opportunities. The adoption of local AI solutions, in particular, could help address the country’s priorities around data security, cost efficiency, and technological autonomy.

Having been present in Vietnam for three decades, HP considers supporting the country’s technological ambitions a key priority. The challenge now for the country will be to harness AI not only as a tool for greater productivity, but also as a catalyst for sustainable growth and long-term competitiveness.