Microsoft and OpenAI have entered a new definitive agreement that fundamentally reshapes their landmark partnership.
The deal allows OpenAI to transition its complex “capped-profit” structure into a Public Benefit Corporation (PBC), a move that provides greater operational flexibility while cementing Microsoft’s role as its key partner in the race toward artificial general intelligence (AGI).
The announcement was met with a positive market reaction, causing Microsoft’s stock to jump 4%.
A New Corporate Structure for a New Era
At the heart of the agreement is OpenAI’s structural evolution. The company will move from its unique capped-profit model to a Public Benefit Corporation, a structure also used by AI labs like Anthropic.
This change is designed to make OpenAI more investor-friendly and simplify its ability to raise capital, potentially paving the way for a future IPO. While becoming a for-profit entity, the new PBC will remain controlled by the original OpenAI nonprofit, ensuring its core mission of benefiting humanity remains central.
Evolving Investments and Commitments
Under the new terms, Microsoft will hold a 27% stake in the restructured OpenAI, with its investment valued at approximately $135 billion. This represents a slight dilution from its previous 32.5% stake but solidifies its position as the largest external investor.
In a massive show of commitment, OpenAI will purchase an additional $250 billion in Azure cloud services from Microsoft. However, in a huge shift, Microsoft will no longer have exclusive rights as OpenAI’s compute provider, giving the AI leader more flexibility.
Greater Independence and the Road to AGI
The restructured partnership grants both companies more independence. OpenAI can now collaborate with third parties, serve U.S. government clients on any cloud platform, and release certain open-weight models. Microsoft also gains the freedom to pursue AGI development independently or with other partners.
Crucially, the deal extends Microsoft’s intellectual property rights to OpenAI’s models through 2032, including those developed after AGI is achieved. In a key governance change, any declaration of AGI by OpenAI must now be verified by an independent expert panel, adding a new layer of oversight to their shared mission.
