Musk-OpenAI Trial Previews How Antitrust Will Steer AI’s Future

The trial between Elon Musk and OpenAI—in which the jury on Monday decided in favor of OpenAI due to the statute of limitations—was about many things: a clash of egos, a governance dispute, and a morality play with few morals.

But it may be remembered as the first antitrust case of the artificial intelligence era, even though antitrust appears nowhere in the complaint.

The antitrust question raised by Musk isn’t whether OpenAI abandoned its founding mission. It’s whether the companies at the frontier of AI will become independent competitors or remain financially, technologically, and operationally dependent on today’s tech giants.

Contrary to Musk’s assertions, OpenAI’s argument for operating a for-profit arm has real merit. Developing cutting-edge AI requires extraordinary investments in chips, cloud infrastructure, data centers, engineering talent, and distribution. OpenAI believes it needs access to private capital—offering investors a return—to compete with firms such as Alphabet Inc.’s Google, Amazon.com Inc., and Meta Platforms Inc.

As a matter of textbook economics, that claim is persuasive. But the need for capital and technology raises a difficult antitrust issue. The firms with the resources to finance AI at scale, or supply it with necessary infrastructure, are the same ones that dominate today’s technology markets.

Microsoft Corp.—OpenAI’s most important strategic partner—has invested roughly $13 billion in the company and holds a significant stake in its for-profit arm, while OpenAI’s models are delivered primarily through Microsoft’s Azure cloud platform. Amazon has committed up to $8 billion to Anthropic and serves as its primary cloud partner, while Google has made multibillion-dollar investments and provides key technology and support.

This is the paradox of AI competition. There are thousands of AI firms, and the emergence of innovative leaders such as OpenAI and Anthropic suggests a dynamic market. Yet these leaders depend on the incumbents they might otherwise displace. The same tech giants that stand to benefit from AI’s growth have the incentive and capacity to ensure it doesn’t disrupt their existing businesses.

To his credit, Musk made this conflict of interest unmistakable. If Musk can’t own or control OpenAI, he’s better off with OpenAI functioning below scale as a pure nonprofit. But this same conflict applies to Microsoft, Google, and Amazon, even if they exert their influence in more subtle ways.

This conflict has precedents, such as Microsoft’s attempt to protect its operating system dominance from the threat posed by Netscape Communications Corp.‘s web browser in the 1990s, but history is a limited guide to AI’s future. As a general-purpose technology, AI’s development will shape myriad markets at once.

If they choose to use them, the antitrust agencies have powerful tools at their disposal. So far, they’ve signaled concern but stopped short of enforcement.

The Federal Trade Commission has opened inquiries into AI partnerships and identified risks associated with equity stakes, revenue-sharing agreements, exclusive cloud arrangements, and governance rights that confer influence without formal control. The FTC hasn’t yet filed a formal complaint that could produce a consent decree with specific commitments and agency oversight. At the same time, the agencies have withdrawn outdated collaboration guidelines without replacing them, leaving uncertainty about how these arrangements will be evaluated.

For the agencies, the policy challenge is straightforward to describe but difficult to resolve. Agencies can’t ignore arrangements that give dominant firms the ability to limit emerging competition in AI. They also can’t simply prohibit those arrangements without risking worse outcomes.

If Microsoft were barred from investing in OpenAI, it might redirect resources toward its own internal AI systems. Cloud providers such as Amazon supply complementary infrastructure that AI firms need to operate and innovate. The development of AI requires the contributions of both existing and new firms.

These concerns extend beyond the most visible AI firms to ones at earlier stages of growth. Recent research has called attention to “killer acquisitions,” in which large firms acquire startups to prevent startups from becoming competitors rather than developing their technologies.

Microsoft’s hiring of Inflection AI Inc.’s leadership and licensing of its technology effectively removed a potential competitor from the market. Amazon’s arrangement with Adept and Google’s deal with Character.AI followed a similar pattern, shifting talent and intellectual property toward incumbent firms.

Again, the FTC investigated these transactions but hasn’t yet brought an enforcement action. Many startups are created with the expectation of being acquired; restricting that exit pathway could reduce incentives to innovate.

Ignoring such deals risks allowing incumbents to quietly eliminate future rivals. How can the FTC spot a “killer acquisition”? Investigations can reveal evidence of specific intent to buy and bury. They also can deter anticompetitive behavior across an industry, but only when the threat of enforcement is real.

Antitrust alone can’t dismantle big tech influence over the trajectory of AI. If society wants independent AI firms, it must consider how they’ll be financed.

Public markets offer one path, though they bring profit-driven pressure that OpenAI and Anthropic are seeking to avoid. Government funding would allow technology to be freely shared, but the inevitable political influence may not be directed toward the public good.

A variety of competing financial models, including public-private partnerships, may be our best hope. Economists have long thought such “divided technical leadership” can bring about beneficial innovation.

The Musk-OpenAI trial won’t determine the future of AI, but it offers an early and uncensored view of the forces that will. AI poses fundamental questions concerning the environment, national security, and work that are outside the scope of antitrust. But antitrust can help determine whether AI will be shaped by competition that benefits all of its users or whether it’s quietly steered by today’s technology giants.

This article does not necessarily reflect the opinion of Bloomberg Industry Group, Inc., the publisher of Bloomberg Law, Bloomberg Tax, and Bloomberg Government, or its owners.

Author Information

Prasad Krishnamurthy is a professor of law at UC Berkeley Law, where he teaches and writes on antitrust and finance.

Interested in writing? Review our author guidelines, and submit pitches to Insights@bloombergindustry.com.