OpenAI and Microsoft’s revised partnership is starting to look less like a breakup and more like a carefully priced reset. The latest detail: OpenAI’s revenue-sharing payments to Microsoft are reportedly capped at $38 billion.
Reuters reported that the cap comes from The Information, citing a person with knowledge of the arrangement. Reuters said it could not immediately verify the report, and that OpenAI and Microsoft did not immediately respond to requests for comment outside regular business hours.
The number matters because the companies had already confirmed the structure. In an official Microsoft post, Microsoft said OpenAI’s revenue-share payments will continue through 2030 at the same percentage, but will be subject to a total cap. The reported $38 billion figure gives that cap a practical shape.
The deal is about optionality
For OpenAI, the obvious benefit is predictability. A fixed ceiling on future payments is easier to explain to investors than an open-ended obligation tied to fast-growing products like ChatGPT, API usage, agents and enterprise deployments.
That is especially important if OpenAI is preparing for a public-market future. Reuters noted that the cap could strengthen OpenAI’s long-term pitch to investors as it works toward a possible IPO. A cleaner payment story does not make the business simple, but it does make one large liability easier to model.
For Microsoft, the trade is different. It keeps revenue sharing through 2030, maintains OpenAI as a major strategic partner, and still holds a license to OpenAI IP through 2032. But the license is now non-exclusive, and Microsoft will no longer pay OpenAI a revenue share for model access through Azure, according to the official agreement summary.
Azure is still central, just not exclusive
The most important strategic shift may be distribution. Microsoft says it remains OpenAI’s primary cloud partner and that OpenAI products will ship first on Azure unless Microsoft cannot, and chooses not to, support the necessary capability.
But OpenAI can now serve all of its products across any cloud provider. That single change moves the relationship from exclusive infrastructure dependency toward preferred-partner status.
CNBC’s reporting shows what that looks like in practice. One day after the revised Microsoft agreement, CNBC reported that OpenAI models would come to Amazon Web Services through Amazon Bedrock, including OpenAI’s Codex agent and a managed-agent service powered by OpenAI.
That is not a small channel expansion. AWS is where many enterprise AI buyers already operate. If OpenAI wants to sell into large companies without forcing a cloud migration, it needs to meet those customers where their data, procurement and security controls already live.
Microsoft is accepting a different kind of leverage
The old version of the partnership gave Microsoft a simpler story: exclusive access, deep Azure alignment and a privileged route into OpenAI’s technology. The new version gives Microsoft something narrower but still valuable: defined economics, long-term IP rights and a huge equity-linked position in OpenAI’s growth.
That may be the more realistic arrangement now. OpenAI has become too large, too capital hungry and too commercially ambitious to behave like a single-cloud dependent. Microsoft also has its own AI products to sell, including Copilot and Azure AI services, which increasingly overlap with OpenAI’s direct enterprise ambitions.
The revised agreement lowers one source of tension by putting numbers and timelines around the relationship. It does not remove competition between the two companies. It just makes the competition easier to manage.
What to watch next
The $38 billion figure, if confirmed, will become a key input for how investors value OpenAI’s future margins. The next question is whether OpenAI can turn broader cloud distribution into enough enterprise revenue to justify its compute commitments.
For customers, the practical takeaway is simpler: OpenAI is becoming less tied to one cloud buying path. Azure will still matter. But AWS, Google Cloud and other infrastructure partners are now part of the strategy instead of side doors.
That is the real shift. The AI platform market is moving from exclusive alliances to negotiated access. The companies with the best models still need distribution, compute and enterprise trust. The cloud companies still want the models. The winners will be the ones that can turn those dependencies into products customers can actually deploy.
