The complex relationship between Microsoft and OpenAI is about to face a new turning point.
Media reports indicate that OpenAI plans to release an &#34;open weights&#34; AI language model as early as next week, marking the company's first launch of an open weights model since GPT-2 in 2019, and the first open model following the exclusive cloud service agreement with Microsoft. This move may pose a new challenge to the partnership between Microsoft and OpenAI.
According to insiders, this new model has reasoning capabilities similar to o3 mini and will be released simultaneously on Azure, Hugging Face, and other major cloud service platforms.
Unlike OpenAI's previous closed weights models, open weights mean that enterprises and government agencies can run the model independently.
<h2>May Allow Enterprises to Customize Models</h2>
The &#34;open weights&#34; model is an intermediate form between open-source and closed-source approaches.
Weights are the way AI models learn and establish connections—certain features or connections are assigned higher weights to reinforce specific information.
An open weights model means its weights are publicly available. This indicates that users can view and modify these weights without retraining the model with new data. Open weights models typically reduce usage costs for enterprises and allow them to customize the model.
However, open weights models are not the same as open-source models. Open-source models usually allow you to view their source code composition and sometimes include their training and weight allocation methods.
Considering OpenAI's consistent practice of not disclosing ChatGPT training data, the final &#34;openness&#34; of the new model will depend on the license accompanying the model and whether OpenAI is willing to provide full access to the model's code and training details.
<h2>Challenges to Microsoft's Exclusive Agreement</h2>
The timing of the model's release is noteworthy.
Currently, Microsoft and OpenAI are renegotiating the contract for the latter's restructuring into a for-profit company, with the revenue-sharing agreement indicating that Microsoft receives a 20% share of the revenue from OpenAI's ChatGPT and API platform, while sharing 20% of Azure OpenAI revenue with OpenAI.
The launch of the &#34;open weights&#34; model may weaken Microsoft's exclusive advantage in the AI business.
As competing cloud service providers can also host the model, some Azure customers may turn to lower-cost alternatives or migrate directly to other cloud platforms.
Earlier this year, Microsoft and other cloud service providers quickly integrated DeepSeek's R1 model, demonstrating the rapid dissemination capability of open models in the market.
Microsoft first invested $1 billion in OpenAI in 2019.
According to the current contract, Microsoft exclusively owns the rights to sell OpenAI software tools through the Azure cloud platform and has priority access to OpenAI's technology
Microsoft has also been designated as the sole computing resource provider for OpenAI, although last year Microsoft allowed OpenAI to independently establish a data center project named Stargate.
Over the past year, the two parties have engaged in fierce debates over a core issue: How much equity in OpenAI should Microsoft receive in exchange for its investment of over $13 billion after restructuring? The discussion has ranged from 20% to 49%.
In addition to the equity battle, several core terms of the current contract are being renegotiated, including Microsoft's exclusive rights to sell OpenAI software through Azure, priority in providing computing infrastructure for OpenAI, and access to AI intellectual property rights before achieving &#34;general artificial intelligence.&#34;
For OpenAI, reaching an agreement with Microsoft is a matter of life and death. The company must obtain Microsoft's approval for its transition to profitability by the end of the year, or it risks losing billions of dollars in investor funding, including from SoftBank

OpenAI

MSFT

MSFL

MSFO

MSFD

MSFU

MSFX

MSFY

Media reports indicate that OpenAI plans to release an "open weight" AI language model as early as next week, marking the company's first launch of an open weight model since GPT-2 in 2019. The introduction of the "open weight" model may weaken Microsoft's exclusive advantage in the AI business. As competing cloud service providers can also host the model, some Azure customers may turn to lower-cost alternatives or migrate directly to other cloud platforms

- OpenAI plans to release an "open weight" AI language model, marking its first such model since GPT-2 and the first after signing an exclusive cloud service agreement with Microsoft.  
- This new model, similar in reasoning capability to o3 mini, will be available on platforms like Azure and Hugging Face, potentially challenging Microsoft's exclusive advantages in the AI sector.  
- The release comes amid ongoing negotiations between Microsoft and OpenAI regarding equity stakes and contract terms, with OpenAI needing Microsoft's approval for its transition to a for-profit entity by year-end to secure significant investments.

Wallstreetcn

Microsoft

GraniteShares 2x Long MSFT Daily ETF

YieldMax MSFT Option Income Strategy ETF

Direxion Daily MSFT Bear 1x Shares

Direxion Daily MSFT Bull 2X Shares

T-Rex 2X Long Microsoft Daily Target ETF

Kurv Yield Premium Strategy Microsoft MSFT ETF

OpenAI makes a significant move: launching the "open weights" model for the first time in six years, is Microsoft's exclusive agreement facing a challenge?