OpenAI may now also use cloud providers other than Microsoft

Microsoft will no longer be the exclusive cloud provider for OpenAI. The partnership between the two companies, which has been in place for several years and runs until 2030, has been extended accordingly. OpenAI is thus granted the right to procure data center capacities elsewhere if required, should Microsoft not wish to become a provider itself. OpenAI CEO Sam Altmann has already stated on several occasions that the lack of computing capacity is a key obstacle to growth for his AI company.

However, Microsoft intends to remain an important investor in OpenAI, according to the press release. Also, unaffected by the change is the fact that the software giant has access to OpenAI's intellectual property for use in services such as Copilot. The Copilot AI chatbot is currently based on ChatGPT 4. Furthermore, the OpenAI API will continue to run exclusively on Azure and will only be available via Azure. The agreements on revenue sharing, which have not been made public, will also continue as before.

500 billion for Stargate

Parallel to the change, OpenAI's participation in a new joint venture called Stargate was announced. A total of 500 billion US dollars is to flow into AI data centers over the next four years. The newly sworn-in US President Donald Trump made the project public at the White House. The software company Oracle and the Japanese technology group Softbank, owned by billionaire Masayoshi Son, are also on board.

Oracle had already announced last June that it would provide data center capacity for OpenAI. As Bloomberg writes, this was still a one-off exception with Microsoft's approval at the time. Microsoft and OpenAI have been closely intertwined since 2019. The software giant has invested billions in the AI start-up several times, with significant parts of the investment being provided in the form of Azure cloud computing power.

(axk)