This article is part of our series that explores the business of artificial intelligence
One of the important highlights of OpenAI’s Spring Update was the release of a native macOS application that can embed ChatGPT into your desktop applications. You can provide ChatGPT with screenshots and copy-paste information into the app and have it reason over it and help you accomplish tasks. It is a very convenient tool that can improve the user experience tremendously.
However, given that Microsoft is OpenAI’s main financial backer, didn’t it make sense for Windows to have the desktop app first?
Users have been speculating on the macOS release, with some calling it a loss for Microsoft. One report indicates that OpenAI prioritized the macOS app “because that’s where more of its users are.” The Windows app will come later in the year.
However, when viewed in the broader scheme of the powerplays in the AI market, it makes sense for Microsoft to not rush on having a native ChatGPT. Microsoft is building its own AI ecosystem around its Copilot brand.
A desktop ChatGPT app would eat into its share of user attention. On the other hand, Microsoft can integrate the same features into its Copilot assistant, with GPT-4o working behind the scenes. At the same time, Microsoft will also have the flexibility to change the backend model to one of its on-device SLMs such as Phi-3, an open source model such as Llama 3, or its own upcoming frontier model MAI-1.
This will help Microsoft keep its AI products loosely coupled with OpenAI. At the same time, it will give Microsoft a foothold into the Apple ecosystem. The native integration of ChatGPT into macOS will increase its usage. And with the model’s running on Microsoft cloud, it means a larger Azure bill for OpenAI and more cash for Microsoft. The same can be said about the reported deal between OpenAI and Apple to integrate ChatGPT to iOS 18 features.
For the time being, this seems to be another win for Microsoft, who has so far made all the right moves in the genAI craze.