Oracle is expanding its AI infrastructure strategy by significantly increasing its use of Nvidia’s latest-generation chips while adding competing technologies from Google and Cerebras Systems. The move reflects the rapidly growing demand for AI computing capacity and highlights how major cloud providers are increasingly balancing scale with chip supplier diversification.

Oracle Expands Nvidia Deployment

Oracle and Nvidia said the cloud company will expand its deployment to “multi-gigawatt scale” using Nvidia’s Vera Rubin AI servers. The companies did not disclose the financial value of the agreement, leaving the precise commercial impact of the expansion unconfirmed.

The scale nevertheless illustrates the infrastructure requirements associated with rapidly expanding AI workloads. Oracle is increasing its access to Nvidia’s newest systems as customers demand greater computing capacity for advanced AI applications, making large-scale data-center infrastructure an increasingly important component of the cloud business.

Vera Rubin Raises the Stakes for AI Infrastructure

Nvidia’s Vera Rubin platform represents the latest generation of systems designed for demanding AI workloads. Nvidia CEO Jensen Huang has previously said that Vera Rubin systems can generate as much as $40 billion in revenue per gigawatt for the chipmaker, providing an indication of the economic scale attached to high-density AI computing infrastructure.

Oracle’s decision to expand its deployment therefore has implications beyond its own cloud capacity. Large orders from cloud providers are an important source of demand for Nvidia and reinforce the broader investment cycle involving processors, servers, networking equipment and data centers.

Google and Cerebras Add a Diversification Layer

At the same time, Oracle is broadening its AI infrastructure portfolio by adding offerings from Alphabet’s Google and Cerebras Systems. The move suggests that Oracle does not intend to rely exclusively on one supplier as demand for AI computing continues to accelerate.

Supplier diversification can give cloud providers greater flexibility in matching different chips and computing architectures to specific workloads. It can also reduce dependence on a single technology platform as the AI infrastructure market becomes more competitive. For Oracle, the combination of expanded Nvidia capacity and alternative offerings creates a broader foundation for serving customers with different performance and computing requirements.

Cloud Competition Moves Toward Computing Capacity

The strategy reflects a wider transformation in the cloud market, where access to advanced AI processors has become increasingly important to competitiveness. Cloud providers are investing heavily in infrastructure while technology companies develop specialized chips to challenge Nvidia’s dominance in AI computing.

Oracle’s next phase will depend on how quickly the expanded Nvidia systems are deployed and how effectively the company integrates Google and Cerebras offerings into its cloud platform. Investors will also be watching whether the multi-gigawatt expansion translates into stronger AI-related cloud demand and revenue. As AI workloads continue to expand, Oracle’s ability to combine large-scale capacity with a diversified chip portfolio could become an increasingly important element of its position in the global cloud infrastructure market.