In 2025, SoftBank led a $40 billion funding round in OpenAI, the American artificial intelligence company based in San Francisco. The round, which became one of the largest private investments in the technology sector, included $30 billion specifically earmarked for compute infrastructure. The deal underscored the escalating capital demands of frontier AI development and the strategic competition between major technology investors.
The investment came during a period of rapid expansion for OpenAI, which had released ChatGPT in November 2022 and subsequently driven the broader AI boom. By the time of the 2025 round, OpenAI had transitioned its corporate structure, with the for-profit subsidiary becoming OpenAI Group PBC, a public benefit corporation 26% owned by the nonprofit OpenAI Foundation. The company was developing increasingly powerful large language models and expanding into areas such as Codex, an AI coding agent.
SoftBank's Strategic Position
SoftBank, the Japanese conglomerate led by Masayoshi Son, had previously made significant bets on technology companies through its Vision Fund. The $40 billion commitment to OpenAI represented a major escalation of its involvement in artificial intelligence. The allocation of $30 billion for compute was notable, as it reflected the understanding that advanced AI models require enormous computational resources, often sourced through Microsoft Azure and other cloud providers.
The investment gave SoftBank a substantial stake in one of the most valuable AI companies worldwide. As of March 2026, OpenAI closed a funding round with a post-money valuation of US$852 billion, making it one of the most valuable AI pure-play companies, rivalled only by Anthropic. The 2025 round contributed to this valuation trajectory.
Compute Investment Details
The $30 billion compute allocation was directed toward expanding OpenAI's infrastructure capabilities. This included contracts with data center operators and partnerships with suppliers of advanced semiconductors. The compute investment was widely interpreted as a response to the intensifying demand for training and running increasingly large models, including efforts toward artificial general intelligence.
OpenAI's reliance on massive computing resources had been a defining feature since its founding. In August 2016, Nvidia gifted its first DGX-1 supercomputer to the organization to help train larger models. By the mid-2020s, the scale of compute required had grown exponentially, with the company partnering with Microsoft, which had invested over $13 billion into OpenAI, and participating in the Stargate Project infrastructure venture with the US government.
Market Context
The 2025 SoftBank-led round occurred against a backdrop of intense competition in the AI sector. Rival firms such as Anthropic, founded by former OpenAI personnel, were also raising substantial capital. Other companies like Thinking Machines Lab, Safe Superintelligence Inc., and SpaceXAI, all founded by OpenAI alumni, added to the competitive landscape.
Investors were increasingly drawn to AI companies due to the commercial success of generative AI products. ChatGPT had become one of the most visited websites globally, and OpenAI's GPT models were widely used across industries. The capital raised in 2025 was intended to maintain OpenAI's leadership position and accelerate the development of more advanced models, including image generation through GPT Image and autonomous agents.
Corporate Governance and Oversight
Under the terms of the 2025 round, SoftBank gained a significant voice in OpenAI's governance, though specific board representation details were not fully disclosed. The investment followed a period of corporate turmoil at OpenAI, including the November 2023 board decision to remove Sam Altman as CEO, followed by his reinstatement five days later after a board reconstruction. The episode highlighted tensions between commercial acceleration and AI safety concerns.
In 2024, roughly half of OpenAI's AI safety researchers left the company, often citing its deprioritization of safety. This exodus included prominent figures such as Ilya Sutskever, who later founded Safe Superintelligence Inc. The 2025 funding round reaffirmed investor confidence in OpenAI's leadership despite these departures.
Compute Strategy and Cloud Partnerships
The compute portion of the investment was linked to OpenAI's ongoing relationships with cloud providers. Microsoft's Azure remained a primary platform for OpenAI's systems, and the company also expanded into other infrastructures. The term "compute" in such funding contexts typically encompasses the procurement of chips, data center space, and energy resources needed to train and run models.
The scale of OpenAI's compute needs had become a subject of industry discussion. Reports indicated that training frontier models could cost hundreds of millions of dollars, with costs rising further for subsequent generations. The $30 billion allocation was designed to secure capacity over multiple years and hedge against potential hardware shortages.
Regulatory and Legal Environment
The 2025 round proceeded amid growing regulatory scrutiny of AI companies. OpenAI had faced multiple lawsuits alleging copyright infringement, including actions by authors and media companies. Several US states had also sued, alleging that ChatGPT coached self-harm and encouraged delusions. These legal challenges did not deter SoftBank's investment, which was structured to accommodate future regulatory changes.
OpenAI's partnerships with defense and government entities also drew attention. The company worked with the Department of Defense, Los Alamos National Laboratory, and arms company Anduril Industries through the Stargate Project. These relationships added a layer of complexity to the investment, as some stakeholders raised questions about the ethical implications of AI deployment.
Broader AI Investment Landscape
The SoftBank-led round was part of a broader wave of massive AI investments. Other notable deals in the same period included Anthropic's multiple funding rounds with Amazon and Google, each reaching billions of dollars. Infrastructure companies such as CoreWeave also attracted significant capital to build specialized AI data centers.
These investments reflected an AI boom era in which startups and established firms alike sought to secure compute, talent, and market share. The $40 billion round set a benchmark for corporate investment in a single AI company, surpassing earlier records. It also signaled SoftBank's conviction that AI would remain a dominant technological and economic force.
Implications for OpenAI's Future
The 2025 funding round provided OpenAI with substantial financial resources to pursue its mission of ensuring that AGI benefits all of humanity. The compute investment was expected to enable more ambitious research projects, including those related to superintelligence. OpenAI's leadership, including CEO Sam Altman and co-founder Greg Brockman, framed the funding as validation of the company's long-term strategy.
However, the round also highlighted ongoing tensions. The sheer scale of compute spending raised questions about sustainability and environmental impact. Safety advocates, including former employees like Andrej Karpathy and John Schulman, continued to call for more cautious development. The 2025 investment thus represented both an opportunity and a challenge for OpenAI as it navigated its dual obligations to investors and the public interest.