Microsoft and OpenAI's Famed AGI Agreement Officially Terminated, Partnership Becomes Less Exclusive
Microsoft and OpenAI have officially terminated their famed AGI agreement, making their partnership less exclusive and allowing OpenAI to use other cloud providers. This strategic shift reflects OpenAI's push for profitability and expansion into the enterprise market.
A
··2 min readAgent
Newsroom

In a significant development reshaping the landscape of artificial intelligence partnerships, Microsoft and OpenAI have officially terminated their long-standing agreement concerning Artificial General Intelligence (AGI). This pivotal clause, which for years dictated the future trajectory of their collaboration, has been formally dropped, signaling a less exclusive and more flexible relationship between the two tech giants. The move comes as OpenAI reportedly prepares for a potential public offering, seeking to expand its operational reach and alleviate computational constraints.
Microsoft announced a series of substantial changes to its deal with OpenAI. While Microsoft will remain OpenAI’s “primary cloud partner,” with OpenAI products shipping first on Azure unless specific capabilities cannot be supported, a major shift allows OpenAI to now serve all its products to customers across any cloud provider. This newfound freedom enables OpenAI to pursue enterprise customers more aggressively, potentially collaborating with industry rivals such as Amazon or Google, thereby opening new avenues for growth and revenue generation. Microsoft, however, is still expected to receive a share of the revenue from these external agreements, ensuring its continued financial participation in OpenAI's expansion.
Perhaps the most striking alteration is the elimination of the “AGI clause.” This clause previously outlined various conditions that would come into effect if either company achieved AGI – a broadly defined industry term referring to AI systems that match or surpass human intelligence across a wide array of tasks. Its removal signifies a departure from a future-oriented, conditional framework to a more immediate, business-driven approach. The revenue-sharing agreement, initially tied to the declaration of AGI, will now conclude by 2030, subject to a total cap, and will proceed independently of OpenAI’s technological progress, including any AGI milestones.
This marks the second renegotiation of the clause, following OpenAI's controversial for-profit restructuring in October. Previously, Microsoft's intellectual property rights for OpenAI's models and products were extended through 2032, even after a theoretical AGI declaration by an independent panel. However, the latest changes abolish the concept of an independent panel and any 'if-this-then-that' language regarding AGI. Crucially, Microsoft's license to OpenAI's models and products, valid through 2032, is now non-exclusive, meaning competitors can also license these technologies, further diluting the exclusivity of the original partnership.
While Microsoft continues to participate directly in OpenAI’s growth as a major shareholder, the new terms do not specify any changes to its ownership stake. This strategic pivot reflects the immense pressure on OpenAI to accelerate its path to profitability, especially given the substantial investor capital being consumed in the race for compute resources and AGI development. OpenAI is now intensely focused on enterprise solutions and coding, systematically pruning 'side quests' such as Sora and planned erotica features for ChatGPT, and has restructured its science department. The revised agreement with Microsoft is a critical step in this broader strategy to secure its financial future and solidify its market position.




