Microsoft and OpenAI have adapted the collaboration agreement they began six years ago to adapt it to a phase of the Artificial Intelligence (AI) market.
Both companies began a collaboration in 2019 launching new computing capabilities in Azure to train large models (LLM) and accelerate the advancement of AI. This Azure supercomputer is responsible for training GPT, the OpenAI LLM that powers the ChatGPT ‘chatbot’and the progress made in it They have also been moved to Microsoft services, such as Copilot.
The two companies have updated the agreement, which will be valid until 2030to adapt it to the new phase of AI that begins, starting with OpenAI stops depending exclusively on the Microsoft supercomputer.
“This new agreement includes changes to exclusivity on new capacity, moving to a model where Microsoft has a right of first refusal (ROFR). To further support OpenAI, Microsoft approved OpenAI’s ability to create additional capacity, primarily for the research and model training”, they explain in a press release.
Microsoft not only maintains the rights to the intellectual property of OpenAI, that is, the model and infrastructure, to implement them in its own products and also make them available to its customers, but it also remains a “major investor in OpenAI”, and both companies have agreed on revenue sharingto ensure that both “benefit from greater use of new and existing models.”
The update of the agreement comes after the president of the United States, Doland Trump, announced a project to create artificial intelligence infrastructure in the country, called ‘Stargate’, which will be in the hands of the ‘joint venture’ formed by OpenAI, SoftBank and Oracle.
#ChatGPT #cuts #umbilical #cut #Microsoft #change #drive #phase
