Oracle admits it entered the cloud AI race later than some competitors, yet underscores how its engineering strengths, high-performance infrastructure, and integrated AI innovations now enable enterprises to leapfrog legacy limitations and accelerate smarter, faster, AI-driven outcomes.
At a time when AI infrastructure, multi-cloud architectures, and enterprise-grade GenAI adoption are accelerating globally, Oracle has announced several significant innovations aimed at redefining data, cloud, and AI strategy for modern enterprises.
Fresh from Oracle AI World, Premalakshmi PR, Vice President – Cloud Infrastructure, Oracle India, shares exclusive insights with CIO&Leader on the company’s latest announcements, from the AI-native Oracle Database 26ai and the new AI Data Platform to the Zettascale10 supercluster, AI Factory, and Oracle’s expanding multi-cloud partnerships.
In this conversation with Jatinder Singh, Editor, CIO&Leader, she discusses what these developments mean for India and APAC, how Oracle is addressing enterprise AI challenges and infrastructure readiness, industry-wise adoption trends, and how CIOs should prepare for the next era of AI-native leadership.
CIO&Leader: Oracle is known for its legacy in databases and ERP, but in cloud AI, hyperscalers like AWS, Azure, and Google lead the way. How is Oracle changing that perception and proving it can compete at the same scale to win customer trust?
Premalakshmi PR: Absolutely, Jatinder. Even if we were the last to enter, we are the latest in terms of technology. We bring all the latest innovations, informed by a decade of learnings from the market and industry. That’s what differentiates OCI, delivering superior price-performance, enhanced security, and innovation at scale.
Oracle has been trusted for decades to manage the world’s most valuable data. Over the last five decades, we have served enterprises, governments, and defense organizations across the globe.
By embedding AI directly into the database layer, at no additional cost, where the intelligence actually lives, we eliminate latency, avoid costly data movement, and enhance advanced analytics. Today, the Oracle Database is truly AI-native.
Our differentiation extends across the stack:
- Infrastructure-level AI/ML: OCI’s global infrastructure itself is powered by AI.
- Data-layer AI: Vector search, Select AI for natural-language insights, and autonomous capabilities that secure, tune, and patch systems automatically. AI is now at the core of data management.
- Open architecture: Partnerships with multiple AI providers, LLMs, and foundation models give customers flexibility to leverage AI across the stack.
Through multi-cloud partnerships, OCI services can be consumed anywhere, on-prem, hybrid, or multi-cloud. The Dedicated Region Cloud@Customer brings nearly 200 cloud and AI services directly into enterprise data centers in as little as three racks.
These innovations allow Oracle to leapfrog what other hyperscalers have done over the last decade. In India, 55–60% of Oracle’s revenue comes from cloud-native, digital-native, and AI workloads. While we may have been a later entrant, we’ve built a compelling value proposition for customers to leverage the latest technology and accelerate their business.
CIO&Leader: At the recent Oracle AI World, you made several announcements, including Oracle Database 26ai, highlighting AI. Could you share how AI is being integrated across data management and what these developments mean for Oracle?
Premalakshmi PR: Over the three days at Oracle AI World, there were several major announcements. Let me start with Oracle Database 26ai. This brings AI into the core of data management, marking a milestone in Oracle’s vision for AI-driven data. It integrates AI across the entire data and development stack, including vector search, data management, AI for development, applications, and analytics, enabling customers to extract insights from raw data while combining private and public datasets. It represents a leap forward in embedding AI natively into database capabilities.
The second announcement was our AI Data Platform, designed to help customers innovate in the AI era. This comprehensive platform connects industry-leading GenAI models with enterprise data, applications, and workflows. It automates data ingestion, semantic enrichment, vector indexing, and integration with GenAI tools, simplifying the journey from raw data to production-grade AI. As part of this initiative, leading global system integrators and consulting firms are collectively investing $1.5 billion in the Oracle AI Platform, training over 8,000 practitioners and building more than 100 industry-specific use cases for the market.
The third announcement was the Autonomous AI Lakehouse, which provides interoperable data access across multi-cloud environments. By combining Autonomous Database capabilities with Apache Iceberg standards, it breaks down data silos and enables analytics and data discovery across multiple clouds and platforms.
We also reinforced our multi-cloud strategy. Last year, Oracle partnered with Azure, GCP, and AWS, and this year we launched Oracle Multi-Cloud Universal Credits. Originally available only on OCI, these credits now allow customers, new and existing, to consume Oracle Cloud services on other hyperscalers, simplifying operations and cloud procurement while maximizing flexibility.
Another innovation is Acceleron, which enhances networking in Oracle Cloud Infrastructure. By combining dedicated network fabrics, converged NICs, and zero-trust packet routing, Acceleron allows faster, more secure workload movement at lower cost while delivering high performance.
We also introduced the Oracle AI Factory, a comprehensive suite of services to help organizations leverage AI and accelerate cloud adoption. It includes AI education, playbooks, runbooks, digital engagement through the Cloud Success Navigator, reference use cases, centers of excellence, and dedicated support from Oracle experts.
On the hardware front, we announced OCI Zettascale10, the largest supercluster with hundreds of thousands of NVIDIA GPUs, delivering 16 zettaFLOPS of peak performance. This flagship cluster, developed in collaboration with OpenAI as part of the Stargate project, is supported by our long-standing partnership with AMD, helping customers scale their AI initiatives efficiently.
CIO&Leader: Many enterprise CIOs still struggle to integrate AI workloads with decades-old on-prem ERP, HCM, and legacy systems. Scaling AI demands massive infrastructure, advanced networking, and mature MLOps practices. Access to Zettascale is limited to select customers, with regional quotas and waiting periods. How is Oracle making large-scale AI more operationally accessible for a broader customer base?
Premalakshmi PR: Oracle Cloud Infrastructure (OCI) Zettascale10 is our largest supercomputer in the cloud. It is designed with extremely low GPU latency across the cluster, enabling high performance and optimal utilization so customers can scale AI workloads effectively.
Enterprises looking to build and deploy AI at scale can either use the Zettascale supercomputer or opt for dedicated AI clusters. From an India data center perspective, dedicated AI clusters are already available and equipped to support enterprise use cases, whether the goal is performance, efficiency, cost optimization, productivity improvement, or operational acceleration. For the Zettascale10, it is 16 zettaFLOPS of peak performance.
Indian enterprises typically approach GenAI in two ways. First, by leveraging AI agents built into Oracle SaaS and applications, which are fully integrated into the stack and allow customers to leapfrog using agentic AI models. Second, by using AI-powered infrastructure at the IaaS level. OCI offers multiple LLMs and foundation models through partnerships with Cohere, Meta Llama, Grok AI, and others. Customers can bring their own data, combine it with public data, fine-tune the available models, and generate deeper insights.
So the opportunity is twofold: leverage agentic AI capabilities in SaaS, and exploit AI-powered infrastructure through dedicated AI clusters. With an open architecture and a wide range of available LLMs, customers can bring their data, build their own models, and drive insights aligned with their use cases and business outcomes.
CIO&Leader: What does the AI Factory mean for India and the APAC market?
Premalakshmi PR: Cloud migration and AI adoption are accelerating rapidly, and enterprises across industries want to harness AI to drive better outcomes. However, keeping pace with evolving technology and building the right skills remains a challenge.
This is where Oracle’s AI Factory comes in. It provides a global framework for upskilling employees, supporting change management, and leveraging proven runbooks and playbooks from Oracle’s global programs. The platform also accelerates cloud migration and AI implementation, helping customers design, test, and scale AI use cases efficiently.
In India, we have two customer excellence hubs, Delhi and Bangalore, where organizations can bring their data, experiment with AI use cases, and co-create solutions with Oracle experts. The AI Factory allows enterprises in India and across APAC to fast-track their AI adoption by combining tools, services, and expertise in a structured, repeatable way.
CIO&Leader: As you scale data centers and GPU AI computing capacity to support large model training and enterprise AI deployments, there are ongoing bottlenecks, land availability, hardware supply, and tight GPU markets. From a customer standpoint, especially those engaging with AF3, how do you ensure they don’t face delays in delivery or time to value while the infrastructure buildout is still catching up?
Premalakshmi PR: I’ll address this from the India market perspective. Today, we have two data centers in Mumbai and Hyderabad. We already host dedicated AI clusters, including large LLM models, in India. From a Gen AI services standpoint, our suite covers vision, language, speech, and text, all integrated into OCI’s cloud and AI offerings.
For the India market, we are sufficiently provisioned to meet current growth and AI demand. Customers who wish to leverage AI infrastructure or services beyond India also have access to global resources. At this point, I don’t see any supply-demand constraints affecting enterprises seeking to use Oracle infrastructure or Gen AI services.
CIO&Leader: With industry clouds becoming a key differentiator in the AI era, which sectors in India are driving adoption for Oracle?
Premalakshmi PR: Our top industry verticals in India are banking and financial services (30% of revenue), ITES/professional services (23%), and public sector (11%). We are also seeing momentum in retail, media and entertainment, startups, digital natives, and manufacturing/automotive. These industries adopt OCI for both Oracle and non-Oracle workloads, cloud-native or migrated from on-prem.
Last year, bookings in India grew 65% year-on-year, with a 35% increase in consumption. We expect this momentum to accelerate further in FY26 with the new AI innovations.
CIO&Leader: Enterprise AI adoption is rapid, but CIOs worry about hallucinations, bias, and explainability. How is Oracle addressing that?
Premalakshmi PR: It starts with high-quality enterprise data combined with public datasets for the right use cases. Properly trained models on AI-powered infrastructure produce more accurate insights. While hallucinations cannot be fully eliminated, this approach delivers successful outcomes, as we’ve seen across industries.
CIO&Leader: What capabilities and mindset will define effective technology leadership, and what trends are emerging in cloud innovation and AI-native services?
Premalakshmi PR: Continuous skilling and hands-on learning are critical. Leaders must understand new technologies daily, break down data silos, and leverage AI for application development, data management, and operational insights. Data and AI are central to economic growth, enabling informed decisions, new products, and deeper insights into consumer behavior. Cloud migration, modernization, and AI adoption remain essential for competitive advantage.
CIO&Leader: Are enterprise customers comfortable with all AI workloads delivered through Oracle? Could centralizing everything pose risks?
Premalakshmi PR: Enterprises operate in a complex world with multiple applications and siloed data. Oracle’s AI innovations across the stack help remove silos, standardize technology, and embed AI capabilities at every layer. Customers leverage existing technology rather than reinventing it, focusing on outcomes rather than managing multiple layers. Integrating Oracle databases, applications, and AI creates a cohesive, AI-infused ecosystem that customers welcome, delivering tangible value and simplifying AI adoption at scale.