The relationship between OpenAI and Microsoft enter a completely new phase Following the formal end of their exclusive cloud agreement, after nearly seven years of close collaboration, the software giant is no longer the sole cloud provider for the company behind ChatGPT, opening the door for the tech giant founded by Sam Altman to work with other platforms such as AWS or Google Cloud.
Although the headline may sound like a breakup, the reality is more nuanced: Microsoft remains the priority partner It maintains a very important position in the infrastructure and commercialization of OpenAI models, but the "lock" that prevented OpenAI from forging deep agreements with other providers has been removed. This move reshapes the AI ​​landscape and has direct implications for companies and startups in Spain and Europe that base their products on these technologies.
Goodbye to exclusivity: what really changes in the agreement
As both companies have acknowledged, the original exclusivity agreement has been terminated and OpenAI will be able to host and offer its AI services on any cloud providerThis means that their models, including the versions they market to enterprise customers, will no longer be necessarily tied to Azure in terms of infrastructure.
At the same time, a new collaboration framework has been signed in which Microsoft continues as a key cloud partnerThe idea is that new OpenAI products will continue to launch first on Azure, provided Microsoft itself deems it viable and is interested in offering those capabilities. It's a clear priority, but it's no longer a technical monopoly.
At the same time, particularly sensitive clauses, such as the one related to so-called Artificial General Intelligence (AGI), are disappearing. The provision that regulated what would happen if either of the two companies achieved AGI first It is no longer in the contract, which reflects how much the technological context has changed and how difficult it is to fit it onto a piece of paper.
The new agreement also eliminates Microsoft's right of first refusal on OpenAI's future computing, granting the latter much greater autonomy to negotiate large deployments with other providers when it comes to new products or infrastructure not linked to the traditional API.
Impact on the cloud: OpenAI will be able to work with AWS, Google Cloud, and others
The most striking consequence for the market is that OpenAI will no longer depend solely on Microsoft's infrastructureIn practice, this allows you to deploy and market your AI models through multiple cloud providers, which is especially relevant for European companies that, due to latency, cost, or regulatory compliance criteria, prioritize specific regions and operators.
This newfound freedom opens the door to deeper agreements with giants like Amazon and Google. In fact, OpenAI has already closed a multi-million dollar deal with Amazon valued at around 50.000 million to launch a persistent memory AI agent platform known as Frontier, an initiative that Microsoft once considered a threat to its former exclusivity.
The restructuring of the contract dissipates that shock and It expressly regulates the possibility of OpenAI collaborating with other actors without triggering a legal battle. Furthermore, the company has secured a significant funding round of approximately $110.000 billion, with participation from Amazon, SoftBank, NVIDIA, and other investors, further solidifying its independent position.
For the European ecosystem, where many technology companies and public administrations are looking to reduce their dependence on a single US provider, this shift could be attractive. It gives them room to... negotiate deployments of OpenAI models on the cloud that best fit with its requirements for data sovereignty, price or sector regulation.
In parallel, OpenAI is advancing its own infrastructure projects, such as Stargate, linked to SoftBank funding and aimed at creating data centers specifically for AI. This strategy points to a long-term search for computational independencealthough Azure will continue to be the backbone of its operation for the next few years.
Key economic factors: investments, revenues and initial public offering
In financial terms, the agreement is also significantly redefined. Over these years, Microsoft has invested tens of billions of dollars in OpenAIMicrosoft's investment has grown from an initial contribution of $1.000 billion in 2019 to a total of around $13.000 billion. Following recent restructurings, Microsoft's stake has stabilized at approximately 27% of OpenAI's for-profit division.
One of the points that has generated the most headlines is the distribution of income. Some media outlets even suggested that Microsoft stopped sharing revenue with its main AI partnerHowever, official statements from both companies point in a different direction. The revenue-sharing agreement remains in place, albeit with an updated structure and clear boundaries.
OpenAI will continue to pay Microsoft a portion of its revenue until 2030, with a total maximum cap of around 20% of turnover derived from certain products, including subscriptions to services like ChatGPT. In return, Microsoft is no longer obligated to pay OpenAI a specific fee for the internal use of its models in its own products, which greatly simplifies accounting between the two.
Another key area is intellectual property. Microsoft will have license on OpenAI models and technologies until 2032This includes any subsequent developments related to AGI that are validated by independent panels. However, this license is no longer exclusive: other major players may sign similar usage agreements, provided they assume the corresponding terms and costs.
Finally, the new agreement clears the way for OpenAI to go public in the coming yearsThe current framework allows the company to go public, something that market forecasts place before the end of 2026, which could make it one of the most significant initial public offerings of the decade.
Context and regulatory pressures in Europe and the US.
The evolution of the alliance cannot be understood without the regulatory context. Over time, as OpenAI gained prominence, Microsoft gradually assumed an increasingly influential role in the company's governance, something that raised eyebrows among competition authorities. Following the internal crisis that resulted in the brief dismissal of Sam Altman, Microsoft secured a non-voting observer seat on OpenAI's board, with access to privileged strategic information.
However, in 2024 the Redmond company relinquished that position, largely due to the antitrust pressures from US and European regulatorsThey were concerned about the combined dominance they could achieve in the generative AI market. The loss of that seat on the board was an early sign that the agreement was going to change.
Added to all this is the legal front. In the United States, OpenAI faces a federal lawsuit brought by Elon Musk, co-founder of the entityThe lawsuit accuses the organization of deviating from its initial non-profit mission and of excessively strengthening its commercial relationship with Microsoft. While the litigation is not the direct trigger for the new agreement, it contributes to the public and political pressure surrounding the company.
In Europe, the new framework is also interpreted in light of European Union AI Regulation and the increasing scrutiny of potential dominant positions in the cloud computing sector. Allowing OpenAI to operate with multiple providers, rather than relying on a single one, better aligns with EU requirements for competition and systemic risk reduction.
Everything suggests that this diversification of partners will be welcomed by European regulators, who have long warned about the dangers of a few companies controlling both the infrastructure and the models that underpin the next generation of digital services.
What does this mean for companies and startups that use OpenAI AI?
For companies building products and services on OpenAI models, the new scenario blends continuity and change. On the one hand, it remains that Edge APIs and more advanced models will remain tied to Azure in much of its distribution, especially in the short term and until 2032, which means that many users will not see a radical change in their daily lives.
On the other hand, eliminating exclusivity and promoting agreements with other providers clearly reduces the long-term risk of vendor lock-in. Companies and developers in Spain, Latin America, and the rest of Europe Multi-cloud architectures may be considered, combining the OpenAI API with other infrastructure services, or even with open-weight models that the company itself or other actors make available for deployment on their own servers.
For AI startups, the message is quite straightforward: It is not advisable to build the whole castle on a single APIThe experience of this agreement, which has gone from a very rigid exclusivity to a much more open model, demonstrates that even multi-billion dollar alliances can change in a matter of a few years, forcing a rethinking of critical technical decisions.
It is expected that, as a result of this shift, providers such as Amazon, Google Cloud, and even European cloud players will intensify their offerings to host workloads linked to OpenAI models or equivalent alternatives. This could translate, in the medium term, into More price competition, better service level agreements and more favorable data portability conditions.
In short, the end of exclusivity between OpenAI and Microsoft does not represent a traumatic break, but rather a reconfiguration that consolidates Azure as the preferred partner, but opens the playing field to other operators and gives OpenAI room to diversify its partnerships, strengthen its funding and move towards a stock market listing, while the rest of the ecosystem adapts to a clearly more fragmented and competitive AI landscape.