Microsoft Corporation is now turning its approach to artificial intelligence (AI) and distancing itself from OpenAI, the company behind ChatGPT, after it has invested nearly $14 billion in the firm. It is raising questions on how the tech giant’s relationship with OpenAI is going to be in the future and what this could mean for Microsoft’s AI strategy.
Microsoft has been developing an integration of internal and external AI models in its Microsoft 365 Copilot product, a report by Reuters said. The reason is to diversify its AI offerings and reduce dependence on OpenAI’s foundational technology. The shift follows after cost and speed became the concerns, especially for enterprise customers using Microsoft 365 Copilot.
A Microsoft representative confirmed the company’s continued collaboration with OpenAI on cutting-edge models, but emphasized it is still crucial to be using different AI models that depend on the product and experience in order to have optimal performance.
Why This Matters For Microsoft
This is part of a broader trend within Microsoft’s business units, which have been adjusting their usage of OpenAI models. For instance, GitHub, acquired by Microsoft in 2018, started to roll out models from other AI players such as Anthropic and Alphabet Inc.’s Google in October as an alternative to OpenAI’s GPT-4.
Insiders have aired their concerns that Microsoft is too reliant on OpenAI for its AI strategy in the past, especially as the investment is as hefty as $14 billion. In this move to increase diversity, there were reports that discussions were held back in October on how it could convert its investment to stock in OpenAI.
In December, the company announced plans to take out a provision that prohibits Microsoft from accessing its highest AI models once AGI is achieved. This goes to show that both of these companies are adjusting the strategy to have more elbow room and mutual benefits in their relationship.
Cost, Speed, Flexibility Woes
The main force behind Microsoft cutting its usage of OpenAI seems to be cost and speed-related. Business users are being increasingly sought in terms of flexible and price-effective options for AI use, to which Microsoft aims to correspond by using internal AI and third-party ones.
From this perspective, the ability to include alternate AI models in Microsoft 365 Copilot is presented as cutting dependence on OpenAI as much as it can afford while still showing the user the latest AI technology can offer.
What Does This Mean for Microsoft’s Partnership with OpenAI?
Despite the reduction in reliance on OpenAI for certain AI applications, the partnership between the two companies remains strong. Microsoft continues to collaborate with OpenAI on frontier models, utilizing a mix of OpenAI’s and Microsoft’s models to provide a diverse range of AI capabilities across its product offerings.
Although Microsoft is diversifying its AI strategy, OpenAI still plays a critical role in powering some of the most advanced AI solutions. However, the shift to include additional models reflects Microsoft’s desire to ensure flexibility and scalability in its AI offerings for business customers.
ALSO READ | OpenAI’s ChatGPT Experiences Unexpected Outage
