Nvidia’s $100 billion AI Bet on OpenAI Sparks Antitrust Debate

- Nvidia unveils $100 billion OpenAI deal for 10 GW AI centers amid lawsuit risks. - Antitrust concerns raise questions about market fairness and competition. On September 22 and 23, 2025, Reuters, Fox Business, and Forbes reported that Nvidia announced a $100 billion strategic investment in OpenAI to develop at least 10 gigawatts of advanced AI data centers. This deal, which marks one of the largest recent commitments to artificial intelligence infrastructure, aims to support OpenAI’s next-generation AI models. Nvidia will phase the investment to align with the deployment of each gigawatt of capacity, and the company expects the first phase to begin operations in the latter half of 2026 using its forthcoming Vera Rubin platform. This collaboration has raised significant antitrust concerns among legal experts and policymakers. On September 23, Andre Barlow, an antitrust lawyer at Doyle, Barlow & Mazard, told Reuters that the deal may fortify Nvidia’s dominance in the chip market while leveraging OpenAI’s leadership in AI development. “This raises significant antitrust concerns,” Barlow stated, warning the partnership could allow Nvidia and OpenAI to solidify joint market control, limiting opportunities for competitors and increasing barriers to entry. Similarly, Vanderbilt Law School antitrust professor Rebecca Haw Allensworth noted the financial interdependence between Nvidia and OpenAI could create incentives for Nvidia to disadvantage OpenAI’s competitors by offering preferential terms, such as better pricing or faster GPU access. In response to these concerns, Nvidia stated it will maintain a neutral approach and affirmed it will continue to prioritize all customers without favoring OpenAI despite the partnership. In addition to the legal scrutiny, the deal highlights the steep costs of scaling artificial intelligence. On September 23, Sarah Kreps, director of the Tech Policy Institute at Cornell University, remarked to Forbes, “The expense of chips, data centers, and power has consolidated the industry to a few firms capable of financing such large-scale projects,” emphasizing the financial concentration this requires. This partnership follows other notable investments aimed at expanding AI capabilities in the tech sector. For example, just days prior on September 18, 2025, Nvidia announced a $5 billion partnership with Intel to develop custom data center and PC products. Meanwhile, Oracle has also been active, reportedly planning to sell $300 billion in computing power to OpenAI over five years for its “Stargate” project, while also negotiating a separate multi-year, $20 billion deal with Meta for cloud computing services.