CMA declares Microsoft-OpenAI partnership safe from merger probe

It’s been a good week for OpenAI

The UK's Competition and Markets Authority (CMA) cleared the relationship after months of scrutiny.

The UK's Competition and Markets Authority (CMA) announced yesterday that Microsoft's partnership with OpenAI does not trigger a formal merger investigation under the Enterprise Act 2002, the nation's anti-competitive practices law.

The decision concludes an inquiry that began in December 2023, amidst concerns that Microsoft's significant investment and close collaboration with OpenAI could stifle competition.

The CMA's investigation centred on whether Microsoft, a major investor with nearly $14 billion injected since 2019, had effectively gained control over OpenAI's commercial policy.

While acknowledging Microsoft's "high level of material influence," the regulator ultimately concluded that this influence did not equate to outright control.

"On the basis of the available evidence…the CMA does not believe that it is or may be the case that the Partnership, in its current form, gives rise to a relevant merger situation," the regulator said [pdf].

"In particular the CMA does not consider there has been a change of control by Microsoft from material influence to de facto control over OpenAI."

The inquiry was initially prompted by Microsoft's deep financial ties, the integration of OpenAI's technology into products like Azure OpenAI Service and Copilot, and Microsoft's pivotal role in Sam Altman's reinstatement as OpenAI CEO in November 2023.

The CMA feared that this control could lead to a "substantial lessening" of competition in the burgeoning AI sector within the UK.

Specifically, the agency was worried that Microsoft could restrict rivals' access to OpenAI's leading models, particularly in markets where AI access is crucial, and Microsoft already holds dominant positions.

Concerns also extended to the accelerated compute market, where OpenAI's potential as a major customer could be impacted.

However, the CMA's investigation revealed that the partnership's dynamics had shifted significantly.

In January 2025, Microsoft renegotiated its cloud computing agreement with OpenAI, moving to a "first right of refusal" model for certain workloads.

This change, along with Microsoft's waivers allowing OpenAI to build additional computing capacity, including a $500 billion datacentre deal with SoftBank, effectively diluted Microsoft's previous exclusive cloud provider status.

"Material aspects of the [Microsoft] partnership have been changing over the course of the investigation," the CMA explained.

Alex Haffner, a competition partner at law firm Fladgate commented on the CMAs decision:

“The decision was reached on purely procedural grounds – whether Microsoft had, through the operational/other changes announced around the time of Sam Altman’s initial departure from the business, achieved a higher level of control over Open AI (at least for merger control purposes) than it had before they took effect. The CMA has concluded no such change of control has taken place and, in particular, Microsoft doesn’t at this time have the ability to control the company’s commercial policies. In that sense, it is relatively straightforward.

“However, against the current narrative of how to regulate Big Tech and, notably disquiet in some quarters as to whether the CMA is being leaned on too heavily by government to change its approach, the decision will no doubt further fan the flames of discontent amongst certain interest groups, a fact evidenced by the recent open letter to the CMA by tech firms and publishers querying that perceived change of motive.”

Judge denies Musk's bid to halt OpenAI's for-profit shift

The CMA’s ruling came a day after a US federal judge rejected Elon Musk's bid to immediately block OpenAI's transition to a for-profit entity, delivering a blow to the billionaire's escalating legal campaign against the AI giant he helped co-found.

The decision [pdf], handed down by Judge Yvonne Gonzalez Rogers of the Northern District of California, denied Musk's request for a preliminary injunction to pause OpenAI's structural transformation.

OpenAI, originally founded in 2015 as a non-profit research lab, has since evolved into a hybrid entity with a capped-profit arm, and is now transitioning into a more conventional for-profit corporation.

Musk had argued that the conversion breaches OpenAI's founding mission to develop AI technology for the public benefit, not for private gain.

His motion for an injunction alleged that OpenAI and its CEO, Sam Altman were engaged in anti-competitive conduct, and violated commitments made to early donors – including Musk himself –and broke antitrust laws by pressuring investors to steer clear of backing rival AI companies.

Judge Gonzalez Rogers rejected all four of Musk's core arguments, writing in her decision that Musk's legal team "failed to meet their burden of proof for the extraordinary relief requested."

The ruling means OpenAI can continue its corporate restructuring process while the case proceeds.

Despite the denial of the injunction, Gonzalez Rogers acknowledged the significance of the case, citing "the public interest at stake and potential for harm if a conversion contrary to law occurred."

As a result, the court agreed to fast-track the broader case, scheduling a full trial for later this year.