Asia is emerging as a global leader in the adoption of artificial intelligence (AI), with a growing number of enterprises now viewing the technology as critical to their operations. However, a new study by Hitachi Vantara highlights a key issue threatening the effectiveness of this AI momentum—weak and fragmented data infrastructure.
According to the company’s latest State of Data Infrastructure Survey, 42% of enterprises in Asia consider AI essential to their business operations, surpassing the global average of 37%. Despite this enthusiasm, the report reveals a worrying gap between AI ambition and the quality of data that supports it.
AI models in Asia currently deliver accurate results only 32% of the time. This shortfall is mainly due to inconsistent and unreliable data. Many organisations report their data is either scattered, unstructured, or not accessible when needed. On average, data is readily available only 34% of the time, which severely limits AI’s ability to generate timely insights.
Security risks and system outages underline the need for resilience
Alongside data reliability, security remains a growing concern. The report shows that 44% of Asian businesses rank data security among their top challenges, again higher than the global average of 38%. This reflects increasing awareness of vulnerabilities in current data systems, especially with the rise of cyberattacks targeting critical digital infrastructure.
The urgency of addressing these issues became clear during the 2024 global IT outage, which exposed the limitations of outdated backup and recovery strategies. Although many companies had backup systems in place, they still suffered prolonged disruptions because their recovery methods were slow, incomplete, or untested. As data volumes across Asia are projected to rise by 123% over the next two years, traditional approaches to data protection are becoming obsolete. In this context, failure to modernise backup and recovery systems could threaten not only data integrity but business continuity as a whole.
Matthew Hardman, Chief Technology Officer for APAC at Hitachi Vantara, pointed out that AI implementation often starts with short-term pilot projects focused on innovation. However, he warned that these efforts must eventually evolve into fully integrated business systems. “Over time, these projects must transition into core business systems, requiring more than just functionality; they need enterprise-grade capabilities like resiliency, data protection, and regulatory compliance,” he said.
Hardman also questioned whether organisations are doing enough to plan for the long term. “Have businesses factored this into their AI roadmaps? Should organisations start adopting a model where every AI project includes an ‘enterprise tax’—a built-in budget allocation to cover the infrastructure and safeguards required to support AI at scale?” he added.
A smarter approach to backup and recovery
As AI use expands, so too do the threats. Cybercriminals are increasingly targeting the data used to train AI models. This includes backup repositories, which may be infiltrated with manipulated data to influence AI decision-making. Meanwhile, ransomware attacks have become more advanced, focusing not only on encrypting files but also destroying backup copies, leaving companies with no means of recovery.
The complexity of operating across hybrid and multi-cloud environments adds another layer of risk. To combat this, companies must develop flexible and scalable recovery strategies. Technologies such as immutable storage, zero-trust security frameworks, and AI-powered backup systems will be critical. These tools can help detect anomalies early, prevent tampering, and ensure swift recovery in the event of disruptions.
Resilience must also extend beyond technical systems. Whether the cause is a cyberattack, a system failure, or human error, organisations need to be prepared. AI-driven observability tools, which track the status and compliance of data in real time, will become increasingly important. Seamless failover strategies that minimise downtime will be essential, particularly as businesses expand their operations across diverse digital ecosystems.
As World Backup Day 2025 approaches, the focus is shifting from simply having a backup to ensuring data is resilient, accessible, and immediately recoverable. Asia’s rapid AI growth presents a powerful opportunity—but without a strong data foundation, that potential may go unfulfilled. Companies that invest in long-term data resilience, rather than just short-term solutions, will be in the strongest position to lead in an AI-driven future.
