OpenAI powers AI with Google Cloud, ending Microsoft's exclusive hold.

In a significant realignment of cloud computing alliances, OpenAI has announced it will utilize Google Cloud's infrastructure to power its artificial intelligence services, including the popular ChatGPT and its API, in multiple countries. This strategic expansion, finalized in May after months of discussion, marks a notable departure from OpenAI's near-exclusive reliance on Microsoft Azure and signals a broader industry trend toward multi-cloud architectures to meet the voracious computational demands of generative AI. The collaboration will see OpenAI leverage Google's cloud resources in key international markets, including the United States, Japan, the Netherlands, Norway, and the United Kingdom, to enhance the global availability and performance of its widely used AI models.[1][2]

The decision to partner with Google is a direct response to the immense and rapidly growing need for computing power required to train and deploy large language models.[3] Since the launch of ChatGPT, OpenAI has experienced an unprecedented surge in user demand, creating significant capacity constraints.[4][5] This has prompted the AI research and deployment company to actively diversify its infrastructure providers to ensure stability, reduce the risk of service disruptions, and gain greater operational flexibility.[6][7] By adding Google to a supplier list that already includes Microsoft, Oracle, and the specialized GPU cloud provider CoreWeave, OpenAI is building a more resilient and distributed foundation for its services.[8][2][9] This multi-cloud strategy is becoming a standard for major players in the AI space, who are finding that reliance on a single vendor can create bottlenecks and supply chain risks.[8][10]

This partnership signifies a pivotal evolution in OpenAI's relationship with its primary backer, Microsoft. For years, an exclusive agreement made Microsoft Azure the sole cloud provider for all of OpenAI's workloads, a deal cemented by Microsoft's multi-billion dollar investments.[11][12] However, that exclusivity ended in early 2025, shifting to a "priority supply" or "right of first refusal" model.[13][1][14] This change, which came after Microsoft itself labeled OpenAI a competitor in the cloud space, opened the door for new collaborations.[8][15] While Microsoft remains a crucial partner, retaining exclusive rights to OpenAI's core APIs for developers building on Azure, the new arrangement reflects a more complex dynamic where the two companies are both collaborators and competitors.[16][4] OpenAI's move to diversify is driven by a need to secure vast amounts of computational resources that may exceed what any single provider can supply, especially as the company reportedly aims to expand its data center capacity to nearly 10 gigawatts in the coming years.[17][18][19]

The deal represents a major victory for Google Cloud as it seeks to bolster its position in the highly competitive cloud market, which has long been dominated by Amazon Web Services (AWS) and Microsoft Azure.[8][16] Securing a high-profile, compute-intensive client like OpenAI lends significant validation to Google's infrastructure and its AI-focused capabilities.[6] A key attraction for OpenAI was likely access to Google's custom-designed Tensor Processing Units (TPUs), specialized chips that can offer superior performance and efficiency for certain AI workloads compared to traditional GPUs.[11][16][17] By providing its advanced hardware and scalable infrastructure to a direct competitor in the AI model race, Google is strategically positioning its cloud platform as a neutral, high-performance option for the entire AI ecosystem.[11][10] This strategy is further evidenced by Google Cloud's existing partnerships with other major AI players, including Anthropic, a startup founded by former OpenAI executives.[20][21]

In conclusion, OpenAI's expansion to Google Cloud is a multifaceted strategic decision with far-reaching implications. For OpenAI, it is a necessary step to manage the explosive growth of its services, mitigate dependency on a single provider, and gain access to specialized hardware, ensuring it can continue to scale and innovate.[6][17] For Google, it is a landmark deal that enhances its credibility and competitive standing in the critical AI infrastructure market.[16][6] This alliance of rivals underscores the immense and non-negotiable demand for computing power in the current AI landscape, a reality that is forcing unprecedented collaborations and reshaping the competitive dynamics between the world's leading technology giants.[22][10] The era of cloud exclusivity for major AI players appears to be over, replaced by a more pragmatic, multi-cloud approach that prioritizes capacity, performance, and resilience above all else.