Microsoft, once the solitary provider for OpenAI’s data center infrastructure, has relinquished its exclusive rights. This dramatic shift follows OpenAI’s announcement of their significant new AI infrastructure engagement, dubbed Stargate, involving multiple heavyweights like SoftBank and Oracle.
Redefined Terms with OpenAI
In parallel to Stargate’s unveiling, Microsoft uncloaked a refreshed agreement with OpenAI. The commitment guarantees Microsoft’s ‘right of first refusal’ on each burgeoning OpenAI cloud computing chunk. This simply means Microsoft can be the first to choose to host OpenAI’s AI workloads, but if any requirement remains, OpenAI can reach out to a competing cloud provider.
OpenAI Bids for More Capacity
A dearth of compute resources had hampered OpenAI’s product release timeline in the past, pushing this tech firm towards seeking other capacity provisions. Microsoft, faced with shareholder pressure, finally allowed OpenAI to negotiate an extra capacity agreement with Oracle last June.
OpenAI’s Future with Microsoft
Whilst Microsoft’s once exclusive partnership with OpenAI faces sea changes, a big part of their initial arrangement extends until 2030. This includes access to OpenAI’s intellectual property, broad revenue sharing agreements and OpenAI’s unique API’s exclusivity. Although this will be impacted if OpenAI can develop AI systems generating profits over $100 billion, in line with their agreed definition. Considering this, OpenAI may consider the possibility of reneging on this agreement in order to secure additional funding from Microsoft.
Exclusivity within Azure
The OpenAI API continues to run exclusively on Azure and is available through the Azure OpenAI Service, implying a direct benefit for Microsoft customers seeking access to frontrunner models. With such advancements, we hope to hear more from both OpenAI and Microsoft regarding future collaborations.
Original source: Read the full article on TechCrunch
