George Kurian: Korea’s AI Momentum at Risk Due to Infrastructure Focus Over Data Strategy
As South Korea accelerates its artificial intelligence (AI) initiatives, NetApp CEO George Kurian, a veteran in enterprise data solutions, warns that many organizations are neglecting a crucial aspect: data management.
“Eighty-five percent of the time in an AI project is spent finding, organizing, and preparing data,” states Kurian, chief executive of NetApp, a leading US-based enterprise data platform provider. During his visit to Seoul to commemorate NetApp’s 25th anniversary in Korea, Kurian told The Korea Herald that companies prioritizing model performance and GPU capacity are “starting in the wrong place,” potentially hindering their AI success.
His message underscores that successful AI implementation hinges more on data usability, accessibility, and governance than raw computing power.
“AI relies on high-quality data to deliver accurate results,” Kurian emphasizes. “The critical question is not just the sophistication of your model, but whether your data is usable across your organization.”
With a global presence in over 40 countries, NetApp is a key enabler of enterprise AI, helping organizations structure and manage the vast datasets required for modern AI models. “We are the data platform that all AI models interact with to predict outcomes,” he explains.
NetApp supports core workloads for hyperscalers like AWS, Microsoft Azure, and Google Cloud, and collaborates with many of the world’s largest banks, chipmakers, and research institutions. In Korea, NetApp serves major telecom companies, financial institutions, manufacturers, automotive companies, and healthcare providers. Since entering the Korean market in 2001, NetApp Korea has expanded its reach to support over 5,000 organizations across various industries.
Kurian highlights that these diverse sectors face similar challenges in scaling AI initiatives.
Overcoming Obstacles: Why AI Pilots Struggle to Scale
“There are two primary challenges,” he explains. “First, the inherent complexity of AI projects, particularly the extensive data preparation required upfront. Second, the difficulty of transitioning from a successful pilot project to full-scale deployment.”
“Scaling requires human change management,” he adds. “Engineers now need to adapt to reviewing code generated by AI, rather than writing all the code themselves.”
While avoiding direct criticism of Korea, Kurian acknowledged the country’s strengths, including rapid technology adoption and innovative public-private partnerships. However, he noted that execution remains a key hurdle.
“Korean companies are generally very adept at adopting new trends quickly,” he states. “The core challenge lies in developing a comprehensive data strategy… data needs to be considered a company-wide asset, not just a departmental one.”
This point is particularly relevant as Korea increases public investment in AI, from national foundation model development to expanding GPU infrastructure. Kurian acknowledges the “encouraging” pace but advises, “It’s time to assess the cost-effectiveness of those investments.”
He recommends three key priorities to move forward effectively. “First, experiment rapidly; the organization that learns the fastest will win,” he advises. Companies also need to align their technology stack with clear business objectives and integrate fragmented internal and external data sources to create a complete and actionable view.
However, his final point reinforces his initial premise.
“A common mistake is approaching AI as an infrastructure problem,” Kurian concludes. “When it’s fundamentally a data problem.”
mjh
