In a surprising move that’s sending ripples through the global tech community, OpenAI has signed a cloud computing deal with Google, even as the two companies are locked in an intense rivalry in the artificial intelligence (AI) space. The agreement, finalized in May 2025 after months of private discussions, enables OpenAI to use Google Cloud’s infrastructure to support its ever-growing computing demands—particularly for training and running large language models like ChatGPT.
The partnership represents a strategic shift for OpenAI, which has traditionally relied exclusively on Microsoft’s Azure cloud services to host its models. Now, amid exponential growth and increasing pressure on compute resources, OpenAI is seeking to diversify its cloud infrastructure base and reduce overdependence on a single provider.
Why OpenAI Turned to Google Despite Fierce Competition
What makes this deal particularly notable is the competitive dynamic between the two companies. OpenAI’s flagship product, ChatGPT, is one of the most widely used generative AI tools on the market, while Google is heavily investing in its own Gemini AI platform, formerly Bard, to dominate the same space.
The idea that two direct competitors would engage in this kind of agreement seems contradictory at first. However, industry insiders explain that the sheer scale of compute required for AI development has forced even rivals to collaborate where infrastructure is concerned.
“The deal underscores the fact that the two are willing to overlook heavy competition between them to meet the massive computing demands,” said analysts from Scotiabank in a note shared with Reuters. “Ultimately, we view this as a big win for Google’s cloud unit.”
They also warned, however, that this could signal deeper concern inside Google about the growing threat ChatGPT poses to its dominance in search and AI innovation.
The End of Azure Exclusivity: A Turning Point for OpenAI
Until recently, OpenAI was contractually bound to host its models exclusively on Microsoft’s Azure Cloud. That agreement dated back to Microsoft’s multi-billion dollar investment in OpenAI, which included a commercial licensing deal allowing Azure to serve as the primary backend for ChatGPT.
However, this exclusivity reportedly ended earlier in 2025. OpenAI’s leadership sought more flexibility in infrastructure sourcing, citing rapid growth and the urgent need for expanded global computing access. The company is now using multiple providers—including Google Cloud, Oracle, and CoreWeave—to support both training workloads and inference operations, which are critical to delivering ChatGPT responses in real-time to millions of users.
Despite the shift, OpenAI and Microsoft continue to work closely. They are reportedly in ongoing negotiations to revise the terms of Microsoft’s existing investment, particularly around the structure of OpenAI’s for-profit arm, which was spun out from its original non-profit foundation in 2019.
Cloud Competition and the Rising Power of Google’s Cloud Unit
From Google’s perspective, the partnership is a major validation of its cloud platform’s capabilities. Google Cloud, the third-largest cloud provider behind Amazon Web Services (AWS) and Microsoft Azure, has been striving to close the gap in recent years.
According to Alphabet’s 2024 financial filings, Google Cloud generated $43 billion in revenue, accounting for over 12% of its parent company’s total income. While Google has always had strong internal AI capabilities through DeepMind and its Gemini AI, this marks the first time OpenAI is tapping into that infrastructure—essentially borrowing computing muscle from its primary competitor.
Industry experts believe that this deal may help strengthen Google Cloud’s enterprise positioning, especially among companies building or hosting AI services that want alternatives to Microsoft and Amazon.
Exploding Compute Needs: OpenAI’s Global Infrastructure Expansion
The deal with Google Cloud is just one part of a broader effort by OpenAI to scale up its infrastructure globally in response to unprecedented AI adoption. Since the public release of ChatGPT in late 2022, the platform has experienced skyrocketing growth. OpenAI has had to grapple with the immense cost and complexity of powering large language models, which involve billions of parameters and require massive GPU clusters.
To address this, OpenAI has made several significant moves:
1. Partnership with CoreWeave
OpenAI signed a multi-billion dollar deal earlier this year with CoreWeave, a specialized cloud provider optimized for AI and GPU computing. CoreWeave has become one of the most important players in the AI cloud race, especially for inference-related workloads.
2. $500 Billion Stargate Project
OpenAI is a key partner in the ambitious Stargate project, a $500 billion initiative backed by SoftBank, Oracle, and G42, aimed at constructing massive AI data centers in the U.S. The project is led by former President Donald Trump and involves building multiple AI-optimized facilities in states like Nevada and Texas over the next five years.
3. Building In-House AI Chips
In a longer-term strategy to reduce reliance on third-party hardware like NVIDIA’s GPUs, OpenAI is reportedly designing its own AI chips. These custom accelerators could offer the company greater control, cost savings, and performance optimization tailored to ChatGPT and its upcoming tools.
4. Collaboration with Oracle and SoftBank
Apart from Stargate, OpenAI is collaborating directly with Oracle Cloud and SoftBank on multiple data infrastructure initiatives. Oracle brings its enterprise-grade infrastructure and high-performance computing (HPC) capabilities, while SoftBank contributes capital and access to Asian markets.
Implications for the AI Ecosystem
This OpenAI–Google collaboration offers a real-world example of how practical needs are reshaping competitive boundaries in the tech industry. As the race to build smarter, faster AI accelerates, no single company—no matter how dominant—can meet the infrastructure demands alone.
The convergence of shared needs is pushing companies that were once exclusive rivals to become pragmatic partners in specific domains.
Key Takeaways:
- OpenAI is no longer solely dependent on Microsoft Azure and is now tapping Google Cloud for extra compute power.
- The deal reflects massive infrastructure demands as OpenAI expands its services and prepares for future releases.
- Microsoft remains a significant investor and strategic partner, though its exclusivity has ended.
- OpenAI’s larger infrastructure strategy includes collaborations with Oracle, CoreWeave, SoftBank, and a proprietary chip initiative.
In a time when AI development is progressing faster than Moore’s Law, cloud computing power has become the core battleground. This deal between OpenAI and Google underscores that the future of AI isn’t just about who has the best model—but also about who can scale, serve, and sustain it for global use.
By securing access to Google’s infrastructure, OpenAI is betting on cloud diversification as a way to protect its momentum and ensure resilience. Whether this sets a new precedent for other AI firms remains to be seen—but one thing is clear: the era of AI “coopetition” has officially begun.
The Information is Collected from LiveMint and The Hindu.
