Tech companies have poured billions into the creation of advanced AI models, while governments are in a race to establish sovereign AI infrastructures.
However, beneath the hype lies a subtle shift that could ultimately decide which companies will thrive.
According to NetApp CEO George Kurian, AI models are set to become increasingly accessible to everyone. What will remain challenging to replicate is a company’s unique data.
“We’ve always believed that, in addition to talent and a deep understanding of your business or clients, data becomes the competitive advantage of an industry,” Kurian shared in an exclusive interview with CNBC-TV18.
AI models are becoming commodities
In recent years, companies have gained access to a growing array of AI models from OpenAI, Google, Anthropic, and Meta. Open-source alternatives have also narrowed the gap with proprietary systems.
This trend means that organizations now have access to similar underlying AI capabilities.
“Everyone has access to the same chips, everyone has access to the same models, and the models are mostly trained on the same data,” Kurian mentioned.
If every enterprise can purchase the same computational infrastructure and license similar AI models, merely adopting AI will not yield a sustainable competitive edge.
Instead, differentiation will stem from how organizations integrate those models with their own proprietary data.
Proprietary data is the real moat
Each organization generates unique data that competitors cannot easily replicate.
Banks gather decades of transaction histories and customer insights.
Hospitals build extensive collections of diagnostic images and clinical records.
Manufacturers accumulate operational and production data from their factories.
Retailers analyze consumer purchasing behaviors across millions of transactions.
Kurian posits that these unique datasets represent the most valuable assets organizations possess in the age of AI.
“Your capability as an organization to deeply understand your business, your clients, and your employees through your data becomes the competitive advantage.”
Business leaders should focus less on which AI model to implement and more on whether their data is organized, secure, and accessible enough to yield meaningful AI insights.
Why enterprise AI begins with data
Generative AI systems are only as effective as the information they can access.
While public AI models possess broad general knowledge, they lack specific insights about a company’s customers, products, operations, or internal processes.
Thus, enterprise AI relies on connecting models with high-quality internal data.
This is where data management companies like NetApp come into play.
Instead of developing foundational models, they concentrate on assisting organizations in storing, governing, protecting, and retrieving enterprise data, ensuring that AI systems can utilize it securely and efficiently.
For many firms, the task of preparing data for AI may be more extensive than deploying the AI model itself.
Real-world AI depends on trusted data
The significance of enterprise data becomes more apparent in sectors where accuracy can directly impact lives.
Healthcare is a prime example.
NetApp has collaborated with organizations like Sankara Nethralaya, a leading eye care institution in India, to manage and leverage substantial volumes of medical data. AI applications built on reliable clinical datasets can assist doctors in analyzing medical images, identifying diseases earlier, and enhancing patient outcomes.
While the AI model itself might be accessible to numerous healthcare providers, the quality, scale, and governance of the underlying clinical data often dictate how effective those systems become in practice.
This principle holds true across various industries—from finance and manufacturing to scientific research.
Kurian also emphasized NetApp’s collaboration with the European Space Agency, where vast data management aids efforts in mapping the universe. Such initiatives rely on partnerships among technology providers, research organizations, and cloud partners to extract insights from enormous datasets.
In every instance, the competitive advantage arises not from the AI model, but from the ability to manage and utilize unique data effectively.
Partnerships matter because no company can do it alone
As AI ecosystems become increasingly intricate, companies are leaning more on partnerships rather than developing every capability in-house.
NetApp has expanded its collaborations with firms like Google Cloud and Nutanix to assist customers in integrating AI into enterprise environments.
Kurian believes this collaborative method mirrors the realities of modern data centers, where organizations typically operate multi-vendor technology environments.
“We’ve always thought that leveraging the power of collaboration and community results in better outcomes for clients,” he stated.
For enterprises, successful AI implementations are likely to hinge on the combination of infrastructure, cloud platforms, security, governance, and domain expertise, rather than depending solely on a single technology provider.
Data governance is becoming as important as AI itself
As organizations increasingly use proprietary information to train and deploy AI systems, concerns regarding data residency, sovereignty, and regulatory compliance are becoming more pressing.
Kurian argues that companies must evolve beyond just storing data within national boundaries.
True data sovereignty means organizations should leverage globally available AI innovations while remaining compliant with local regulations and retaining full control over sensitive information.
This capability is particularly vital for governments, banks, healthcare providers, and other highly regulated sectors.
Winning the AI race will depend on who uses data best
Much of today’s discourse around AI revolves around which company has developed the largest model or secured the most advanced chips.
Kurian believes this viewpoint neglects the aspect most likely to influence long-term success.
As AI models grow cheaper, faster, and more accessible, they will become increasingly interchangeable.
What cannot be easily replicated is decades of proprietary enterprise knowledge captured in both structured and unstructured data.
Companies that prioritize organizing, protecting, and understanding that data will be better equipped to build distinct AI applications, enhance customer experiences, and make informed business decisions.
The AI era may be fueled by models, but its leading players will likely be defined by their data strategies.
For enterprises crafting their AI strategy, this might be the most critical insight of all.
