OpenAI has completed its long-awaited recapitalization, restructuring into a new corporate model that splits its operations between a non-profit foundation and a for-profit public benefit corporation (PBC). This marks the culmination of a complex legal and financial transformation that had been fiercely contested, particularly by former co-founder Elon Musk, who opposed the shift and even offered $97.4 billion to acquire the company. Under the new structure, the OpenAI Foundation, a non-profit entity, will control the for-profit OpenAI Group PBC, which is now free to raise capital, pursue growth, and acquire companies without the legal constraints that previously limited OpenAI’s operations. The Foundation will hold a 26% stake in OpenAI Group and retain the right to receive additional shares through a warrant if the company continues to grow. Microsoft, a key early investor and strategic partner, will own approximately 27% of the for-profit entity, valued at around $135 billion. The remaining 47% is held by employees and other investors. This new structure was made possible in part by a $30 billion investment from SoftBank, which was contingent on OpenAI’s successful transition to a for-profit model. The agreement strengthens Microsoft’s longstanding partnership with OpenAI. Microsoft’s intellectual property (IP) rights to OpenAI’s models and products are now extended through 2032, and will continue to apply to models developed after Artificial General Intelligence (AGI) is declared—provided they meet safety standards. Microsoft’s IP rights include confidential research methods used in model development, but exclude model architecture, weights, inference code, and data center hardware and software. Notably, Microsoft no longer has exclusive rights to provide OpenAI’s compute services, though OpenAI has committed to purchasing an additional $250 billion in Azure services. A new provision requires that if OpenAI ever declares it has achieved AGI, that claim must be verified by an independent expert panel. This adds a layer of accountability to the company’s most ambitious goal. Microsoft can now independently pursue AGI development, either alone or with third parties, and if it uses OpenAI’s IP before AGI is declared, those models will be subject to strict compute thresholds—far larger than current systems. The deal also allows OpenAI greater flexibility: it can now jointly develop products with third parties, release open-weight models that meet capability criteria, and provide API access to U.S. government national security customers regardless of cloud provider. Microsoft will retain exclusive rights to API products on Azure, but non-API products can be hosted on any cloud. The changes were influenced by legal scrutiny from state attorneys general in California and Delaware, as well as public pressure. OpenAI chairman Brett Taylor acknowledged these inquiries helped improve the final structure, stating the company is now better aligned with public interest. In response, CEO Sam Altman announced a public livestream with chief scientist Jakub Pachocki to address questions from the public, underscoring OpenAI’s commitment to transparency. The new model aims to balance innovation with responsibility, allowing OpenAI to scale while maintaining its mission to ensure AI benefits humanity. With Microsoft’s continued support and a more flexible legal framework, the company is positioned to advance AI development at an unprecedented pace—while navigating the complex legal, ethical, and commercial challenges of the age of artificial intelligence.