Microsoft Accused of AI Supply Manipulation to Inflate ChatGPT Costs
Antitrust Allegations Surface Against Microsoft in AI Compute Market
Microsoft is facing a significant class-action lawsuit alleging a deliberate strategy to manipulate the supply of artificial intelligence compute resources, thereby artificially inflating the costs associated with services like ChatGPT. The suit, filed by eleven ChatGPT Plus subscribers, contends that Microsoft leveraged its dominant position in cloud computing, specifically through its Azure platform, to exert control over OpenAI's access to essential computational power. This alleged anticompetitive conduct, which the plaintiffs claim persisted for nearly three years, is said to have resulted in overpaid subscriptions and a degraded user experience.
The Core of the Allegations: An Exclusive Cloud Agreement
At the heart of the lawsuit is Microsoft's 2019 agreement with OpenAI. This deal reportedly granted Microsoft exclusive rights to supply the Azure compute resources necessary for OpenAI's commercial models. Plaintiffs argue that this arrangement effectively placed Microsoft in control of a "horizontal competitor's supply chain" while the tech giant was concurrently developing its own generative AI products. The lawsuit characterizes this as a continuation of Microsoft's historical antitrust playbook, drawing parallels to the landmark "U.S. v. Microsoft" case from the 1990s and labeling the company a "recidivist violator" in the burgeoning AI sector.
Defining a New Antitrust Market: The Consumer Generative AI Market
The plaintiffs have introduced a new antitrust market definition: the "Consumer Generative AI Market." This encompasses subscription-based generative AI products, including but not limited to ChatGPT Plus, Claude Pro, Gemini Advanced, and DeepSeek Chat. By defining this specific market, the lawsuit aims to highlight the competitive landscape and the alleged impact of Microsoft's actions on consumer access and pricing within this domain.
Financial Stakes and Alleged Dual Profit Motive
Further complicating the allegations, the lawsuit claims that Microsoft holds a substantial 49% stake in OpenAI's for-profit arm and benefits from 20% of its paid-product revenue. This structure, according to the plaintiffs, allows Microsoft to "profit twice"—first from the sale of compute resources and second from the AI products it allegedly constrains. This alleged dual profit motive is presented as a key driver behind the anticompetitive behavior.
The "Natural Experiment" of Shifting Cloud Providers
A pivotal piece of evidence presented in the complaint is the dramatic shift in ChatGPT's pricing that occurred in June 2025. When OpenAI reportedly began sourcing compute resources from Google Cloud, thereby ending Microsoft's exclusivity, ChatGPT token prices are said to have plummeted by 80% within weeks. The plaintiffs describe this event as a "powerful natural experiment," asserting that it unequivocally demonstrates Microsoft's alleged "anticompetitive restraint." Prior to this shift, users reportedly experienced "poor quality, unreleased innovations, and slow response times," which the lawsuit attributes to Microsoft's alleged throttling of compute supply.
Expert Analysis on Proof and Control
Navodaya Singh Rajpurohit, legal partner at Coinque Consulting, commented on the case, suggesting that the exclusive agreement itself would serve as the strongest evidence. "If that document shows Microsoft held or used control over OpenAI compute, that is primary proof," Rajpurohit stated. He added that internal emails and capacity records could also be crucial if the agreement is not fully accessible. Rajpurohit emphasized that the strength of the claim hinges on whether Microsoft "actually exercised or clearly held that control," noting that the case would be weaker if such control was not demonstrated.
Lingering Contractual Power and Demands for Relief
The lawsuit further alleges that Microsoft "still retains the contractual ability to restrict OpenAI's compute purchases." This retained power is described as a persistent threat, akin to a "sword of Damocles over OpenAI, wielded by one of its principal competitors." In seeking relief, the plaintiffs are requesting monetary damages for the alleged overcharges incurred between November 2022 and February 2025. Additionally, they are seeking a permanent injunction to bar Microsoft from entering into exclusive compute deals with OpenAI in the future. The plaintiffs also demand disclosure of internal communications pertaining to compute supply, pricing, and integration strategies.
Potential Court Interventions and Future Implications
Rajpurohit suggested that the court
AI Summary
A new class-action lawsuit filed in San Francisco federal court accuses Microsoft of anticompetitive practices by allegedly manipulating the supply of AI compute resources to OpenAI. The plaintiffs, eleven ChatGPT Plus subscribers, claim that Microsoft used its exclusive cloud agreement, initiated in 2019, to control OpenAI's access to Azure compute. This alleged control allowed Microsoft to artificially inflate the subscription costs for ChatGPT and other OpenAI services, while simultaneously developing its own competing AI products. The lawsuit posits that this strategy resulted in degraded service, including poor quality, delayed innovations, and slow response times, for users between November 2022 and February 2025. Evidence cited in the complaint includes a significant 80% drop in ChatGPT token prices within weeks of OpenAI beginning to purchase compute resources from Google Cloud in June 2025, an event the plaintiffs describe as a "powerful natural experiment" demonstrating Microsoft's alleged restraint of trade. The legal action seeks monetary damages for alleged overcharges and a permanent injunction to prevent future exclusive compute deals between Microsoft and OpenAI. Legal experts suggest that the exclusive agreement itself would be primary proof, with internal communications and capacity records serving as secondary evidence. The lawsuit also highlights that Microsoft allegedly retains the contractual ability to restrict OpenAI's compute purchases, posing an ongoing threat. The case draws parallels to historical antitrust battles involving Microsoft, characterizing the company as a "recidivist violator" applying an "exclusionary playbook" to the AI domain. The plaintiffs are also seeking disclosure of internal communications regarding compute supply, pricing, and integration, and are advocating for court-ordered changes to the OpenAI arrangement, including the removal of exclusive terms and the establishment of guardrails against future choke points. Microsoft and OpenAI have not yet responded to requests for comment.