Microsoft, OpenAI Restructure AI Partnership

Microsoft and OpenAI announced they are redefining their partnership as part of a major recapitalization effort aimed at preparing for the arrival of artificial general intelligence (AGI). Under the new structure, OpenAI's for-profit business will become a public benefit corporation (PBC) called OpenAI Group, valued at about $500 billion, with Microsoft holding a 27 percent stake worth roughly $135 billion. The reorganization gives the nonprofit OpenAI Foundation greater control over the for-profit entity to ensure continued alignment with its mission to advance safe and beneficial AGI development.

A PBC is a type of company that, by law, must balance profitability with a stated mission to create a positive impact on society or the environment.

The nonprofit arm, OpenAI Foundation, will retain control of the for-profit venture, per OpenAI's announcement. The restructuring is meant to ensure that the nonprofit foundation "has a direct path to major resources before AGI [artificial general intelligence] arrives," OpenAI said.

Microsoft retains "exclusive IP rights and Azure API exclusivity" until OpenAI — alongside a panel of independent experts — determines it has achieved AGI. However, under the terms of the recapitalization, Microsoft's exclusivity rights have been extended until 2032, with added guardrails for AGI development.

Additionally, Microsoft's rights to OpenAI's confidential research methods — the proprietary techniques used to develop AI models and systems — will remain until either 2030 or the achievement of AGI, whichever comes first.

There are other new stipulations. Microsoft can pursue AGI either alone or with partners outside of OpenAI. However, "if Microsoft uses OpenAI's IP to develop AGI, prior to AGI being declared, the models will be subject to compute thresholds."

Likewise, OpenAI can now develop "some" products with companies other than Microsoft, but API products developed will be exclusive to Azure.

OpenAI has also committed to buying $250 billion in Azure services, though it can also choose to partner with other hyperscale cloud providers; Microsoft will no longer have the "right of refusal."

OpenAI can now let U.S. national security agencies use its tools and services, even if they're not hosted on Microsoft's cloud. It also can publicly release certain versions of its AI models, as long as they meet specific safety and performance standards.

"As we step into this next chapter of our partnership, both companies are better positioned than ever to continue building great products that meet real-world needs, and create new opportunity for everyone and every business," Microsoft said.

For more information, read the Microsoft blog post.

About the Author

Ammaarah Mohamed ([email protected]) is the editorial assistant of the Converge360 Enterprise Technology group.


Featured

  • cable management under a table

    Infrastructure First: Why the Hidden Layer of Your Learning Space Is Your Most Important Technology Investment

    Campuses are spending big on AV technology and wondering why it still doesn't just work. The answer isn't better gear. It's a better foundation.

  • abstract smartphone translucent screen displaying AI interface

    Apple Introduces Redesigned Siri AI

    At its recent Worldwide Developers Conference, Apple introduced Siri AI, a redesigned version of its voice assistant that Apple describes in its own announcement as "a profoundly more capable and personal assistant." The update is intended to make Siri more conversational, more context-aware, and more useful across iPhone, iPad, Mac, Apple Watch, and Vision Pro.

  • person typing on a touch screen schedule plan calendar

    DOJ Extends Deadline for ADA Title II Compliance

    Institutions working to meet the Americans with Disabilities Act Title II regulations for digital accessibility have received a temporary reprieve: The United States Department of Justice has published an interim final rule to push back the compliance deadline by one year.

  • digital data protection and cyber security

    White House Launches New AI Security Framework

    President Donald Trump has issued a new executive order aimed at maintaining United States AI leadership while addressing the security risks posed by increasingly powerful AI systems.