Balancing Trade-Offs in Hybrid Cloud and the Infrastructure Behind Scalable AI – with Jason Hardy of Hitachi Vantara

This interview evaluation is sponsored by Hitachi Vantara and was written, edited, and revealed in alignment with our Emerj sponsored content guidelines. Learn extra about our thought management and content material creation providers on our Emerj Media Services page.

Organizations throughout numerous industries are making vital investments in enterprise AI capabilities to reinforce their effectivity and achieve a aggressive edge in their respective markets. However, many face substantial hurdles, together with integrating AI into current methods, addressing information high quality points, and aligning AI initiatives with enterprise aims.

These challenges typically result in elevated prices and delayed returns on funding. Recent research present {that a} substantial quantity of AI tasks fail to satisfy expectations: One report by S&P Global Market Intelligence discovered that 42% of companies deserted most of their AI initiatives in 2025, up from 17% the earlier 12 months. 

Similarly, an MIT study revealed that 95% of generative AI (GenAI) deployment efforts fail to realize desired outcomes. These statistics underscore the significance of strategic planning and execution in the adoption of AI.

On a current episode of the ‘AI in Business’ podcast, Emerj Editorial Director Matthew DeMello sat down with Jason Hardy, Chief Technology Officer of AI at Hitachi Vantara, to debate how organizations can undertake AI strategically, balancing infrastructure, information readiness, cloud use, compliance, and long-term ROI.

This article brings out two important insights each group wants for efficiently adopting and scaling AI:

• Balancing cloud and information heart use: Running core AI in information facilities whereas utilizing the cloud for low-priority duties controls prices, boosts ROI, and helps sustainability.
• Reframing ROI in AI for long-term success: Valuing classes from failures alongside wins and adopting a long-term, three-year view, the place persistence builds the basis for aggressive benefit.

Guest: Jason Hardy, Chief Technology Officer of AI at Hitachi Vantara

Expertise: Artificial Intelligence, Generative AI, Cloud Computing

Brief Recognition: Jason leads the growth of Hitachi Vantara’s AI technique and portfolio. With over 20 years of expertise in data-driven expertise, Jason has beforehand served as the CTO for Hitachi Vantara’s Data Intelligence portfolio. Jason can be a seasoned marketing consultant, having suggested quite a few world shoppers on information technique, and is a frequent speaker at trade occasions.

Balancing Cloud and Data Center Use

Jason opens the dialog by explaining the dilemma firms face when adopting AI infrastructure: whether or not to commit upfront or begin step by step and totally. Initially, there was a rush to purchase GPUs and not using a clear plan; nevertheless, the strategy is now extra methodical. 

Jason emphasizes to the enterprise viewers that it’s not nearly GPUs, he says; organizations should take into account the place their information comes from, how methods combine, and what outcomes they goal for. These complexities typically grow to be clear solely halfway by means of the journey. 

Additionally, GPUs demand vital energy, elevating sustainability issues till the expertise turns into extra environment friendly. Many companies are adopting hybrid methods, combining on-premises and cloud sources to strike a stability between efficiency, value, and ROI, with out overspending or compromising sustainability.

He additionally emphasizes that implementing AI requires a structured and strategic strategy, moderately than dashing in and not using a plan. Companies can’t merely develop information heart house or deal with AI as a plug-and-play product; they aren’t one thing you “purchase” and immediately activate. 

Instead, he believes that AI adoption is an enterprise journey that requires planning throughout bodily infrastructure, information administration, and expertise. 

Fundamentally, Jason advises enterprise leaders on the must align execution with enterprise technique, contemplating ESG influence, hybrid cloud methods, and general ROI. The period of experimenting to see what sticks is over, he notes emphatically. Boards and CEOs demanding AI should perceive that success comes from deliberate planning and phased execution, not fast fixes.

Jason additionally explains that companies initially turned to the cloud as a handy house for AI experimentation, and it stays a great possibility for particular workloads. However, the cloud works finest for lower-priority, horizontal use instances, corresponding to HR or finance, the place flexibility issues greater than pace.

The drawback is value management; cloud bills rise rapidly as person exercise will increase, and firms can’t simply restrict utilization with out shutting down providers:

“The ROI begins to erode for those who’re not getting a excessive quantity of worth out of your cloud platform, however you’re paying a substantial quantity of value on your per token, or nevertheless you’re paying for it.

That’s why we’re beginning to see a bit of repatriation, bringing information again into the information heart, and then utilizing the cloud extra strategically. In the brief time period, leaders will run the information heart as a result of it’s a finite value.

They need the funding to be one-and-done, and then, clearly, energy and cooling nonetheless matter. So they’ll use the cloud for that fast burst, or use the cloud for low precedence property as a matter of technique, as I mentioned.”

— Jason Hardy, Chief Technology Officer of AI at Hitachi Vantara

Reframing ROI in AI for Long-Term Success

Jason explains that firms are adopting a hybrid strategy, utilizing the cloud for bursts, low-priority duties, or particular strategic wants—whereas transferring core workloads again on-premises. 

The shift, Jason says, allows enterprises to entry delicate information domestically for compliance and safety causes, whereas delivering real-time outputs, which is essential in industries corresponding to manufacturing and vitality, the place latency from cloud round-trips is just too gradual. He notes that generative and energy-focused AI functions demand this on-site functionality. 

At the similar time, he feels geopolitical and regulatory pressures are accelerating “sovereign AI” initiatives, the place nations construct in-country AI infrastructure to keep up management over essential expertise.

Jason’s examples of these sorts of initiatives embody government-backed investments in the Middle East and Japan. These efforts symbolize repatriation “on steroids,” aiming to create cloud-like experiences inside nationwide boundaries whereas preserving autonomy, compliance, and cultural alignment.

Hen continues by saying that ROI in AI can’t be measured in the conventional sense of chopping prices or lowering headcount. Instead, it must be reframed to account for studying by means of failure. Most AI tasks — round 90% — fail or by no means make it to manufacturing, however these failures solely generate insights that enhance the probabilities of success in future initiatives when enterprise leaders study from their errors: 

“What we’ve additionally discovered is that roughly 86 p.c of pilots that succeed show that the juice is value the squeeze.

Whereas with outright failure, all these tasks that gained’t see the gentle of day, we’re seeing the enhancements from these tasks that do make the gentle of day make up for the distinction in funding.

So, there’s a actual profit to AI for the sake of understanding your group higher than ever. But once more, it’s that stability once more, and leaders must make judgments appropriately in opposition to enterprise targets.”

– Jason Hardy, Chief Technology Officer of AI at Hitachi Vantara

He stresses that AI investments ought to be evaluated over a three-year horizon, moderately than primarily based on short-term outcomes. AI is a aggressive benefit, so if opponents succeed first, they’ll reap the rewards. ROI ought to issue in each successes and the classes from failures. 

Companies should settle for some “burn” alongside the approach as a substitute of abandoning tasks too rapidly, as a result of persistence, even by means of failures, finally results in payoff, making the effort worthwhile.