Omar Akar, Area Vice President at Everpure, explains why unified data, resilient infrastructure and strong governance will determine whether organisations can move AI from experimentation to measurable business value.
Enterprise AI is reshaping infrastructure priorities across the Middle East, Central and Eastern Europe, Turkey, CIS and Africa. Organisations are looking beyond compute capacity to address the quality, availability, governance and protection of the data powering increasingly complex AI workloads.
For channel partners, this transition creates opportunities to help customers modernise legacy environments, unify fragmented data and improve GPU utilisation while meeting sovereignty, privacy and regulatory requirements. Cyber resilience has also become central to the conversation, particularly as agentic AI systems begin using enterprise data to make decisions and take action.
Speaking to Tahawultech.com, Omar Akar, Area Vice President, Everpure, discusses the foundations required to move AI from experimentation into production. Akar also explains why unified data strategies, outcome-led infrastructure and trusted partner expertise will determine whether organisations can translate AI investments into measurable business value.
Interview excerpts
How is enterprise AI reshaping data infrastructure priorities across your region?
Infrastructure decisions traditionally centred on application performance, capacity and anticipated growth. Enterprise AI has broadened that discussion because data now feeds expensive compute resources, making data readiness critical to both performance and the economics of AI. How quickly and consistently an organisation can discover, prepare and deliver data directly affects the value of its investment. An idle GPU represents a costly resource waiting for information. Even substantial compute capacity can deliver limited value if the supporting data infrastructure cannot keep pace. Organisations therefore need a data strategy that connects infrastructure decisions with business outcomes. Flexibility also matters because AI workloads can evolve in ways that are difficult to forecast.
Infrastructure must respond to changing requirements while delivering predictable service levels.
What prevents organisations from turning AI experimentation into measurable business value at scale?
The obstacle is often the data strategy. Organisations can invest in GPUs, models and applications, yet still struggle to integrate AI into everyday operations because information remains fragmented, difficult to prepare or poorly governed. A successful experiment may use a carefully selected dataset in a controlled environment. Expanding that application across the business requires reliable access to information from multiple sources, clear ownership and confidence in how the data can be used. Preparation and governance become more demanding as the scope grows. Success depends on unifying data, establishing trust and making information available to applications consistently. Organisations also need to connect AI initiatives with specific operational priorities. Strong data foundations help promising experiments develop into applications that deliver measurable business value.
How can enterprises modernise legacy data environments for AI workloads?
Modernisation should create a foundation for continuous innovation. Organisations need infrastructure that can evolve with business requirements while reducing the disruption associated with periodic technology refreshes and data migrations. Adaptability is particularly important for AI because future performance and capacity needs can be difficult to predict. Infrastructure governed by agreed service levels can help customers manage that uncertainty and focus on the outcomes their applications require. Unification is another priority. Fragmented environments make information harder to discover, prepare and deliver, limiting the data available to AI applications. Enterprises should consider modernisation across the full data lifecycle, including protection, governance and preparation.
A simpler operating model can also reduce the management burden on technology teams.
What role does a unified data platform play in GPU utilisation, application performance and AI economics?
A unified data platform helps connect the stages that determine whether information is ready for AI. Discovering and consolidating data, protecting it, establishing trust and preparing it for applications all affect how effectively compute resources are used. Delays anywhere in that process can leave GPUs waiting for information and slow application performance. Consistent data delivery helps organisations make better use of expensive infrastructure and supports more predictable operations. Governance matters alongside speed. Organisations need to understand what information is available, whether its use is permitted and what preparation is required before it enters an AI application. Our broader direction at Everpure is to address those requirements across the data lifecycle. Storage provides an essential foundation, while the wider platform must help customers turn data into usable, trusted information.
How can organisations accelerate AI adoption while meeting data sovereignty, privacy and regulatory requirements?
Data sovereignty, privacy and regulatory compliance need to form part of the AI strategy from the beginning. Organisations must understand where information resides, how it is governed and whether it can safely enter an AI pipeline. Clear visibility provides the context needed to make informed decisions. Customers need to identify sensitive information, understand applicable restrictions and maintain control over its use across different environments. Confidence in those controls becomes increasingly important when moving from experimentation into production. Addressing governance early can reduce uncertainty and help organisations progress with greater confidence. A trusted data foundation supports AI adoption by making compliance and control part of how applications are developed and operated.
Why should cyber resilience and data protection form part of every enterprise AI strategy?
AI increases the importance of enterprise data and makes its integrity more consequential. Compromised or altered information can affect an application’s output, leading to unreliable recommendations or inappropriate actions. The concern grows as organisations introduce agents capable of acting on information. Confidence in the underlying data becomes essential when systems move beyond generating responses to participating in business processes. Protection must therefore address both integrity and recovery. Organisations need to safeguard information and retain the ability to restore trusted data following an incident. Rapid recovery helps maintain critical operations and limits disruption to applications that depend on that information. Everpure has integrated protection and recovery capabilities into its data platform and service-level commitments.
Our aim is to help customers build resilience into the foundations supporting their AI initiatives.
Which industries are leading AI adoption across the region, and what can others learn from them?
Governments are playing a leading role, particularly in the Gulf, where AI forms part of broader national transformation strategies. We also see growing adoption across telecommunications, healthcare, banking and other financial services. A key lesson is to approach AI as an organisation-wide transformation. Successful adoption depends on how data, processes, technology and cyber resilience work together. Introducing an application is one step; integrating it into operations requires a wider foundation. An organisation’s own information, together with the context surrounding it, can become a significant competitive advantage. Customers need to understand how that data can support their priorities and where AI can deliver practical value. Clear objectives and trusted information provide a stronger basis for scaling adoption.
What will define successful AI infrastructure strategies over the next three years?
Successful strategies will focus on integrating AI into operations and measuring its contribution. Organisations will need to assess whether investments improve processes, deliver a return and create tangible business outcomes. A unified data strategy will sit at the centre of that work. Access to compute, models and applications must be supported by the ability to discover, govern, prepare and deliver information effectively. Infrastructure will also need to adapt as workloads and business requirements change. Simplicity, resilience and predictable service levels will become increasingly valuable. Customers need an operating model that allows technology teams to support innovation without adding unnecessary management complexity. The strongest strategies will connect those infrastructure capabilities with clearly defined business priorities.
What does it mean to you personally to contribute to the region’s AI ambitions?
I feel proud to contribute at a time when the UAE, Saudi Arabia and the wider Gulf are pursuing ambitious plans for AI. Supporting that transformation gives our work a purpose that extends beyond individual technology deployments. Our goal is to grow with the region and help organisations build the foundations required to achieve their ambitions. Data infrastructure plays an important role because it supports how information is prepared, protected and used to deliver business value. I also feel a personal responsibility to contribute. Helping customers strengthen their capabilities and make progress with AI connects our work to broader national priorities. Being part of that journey is both meaningful and motivating.
Source: Tahawul Tech