Microsoft launched three AI models MAI without OpenAI — a signal of technological independence

Microsoft has released three proprietary AI models — MAI-Transcribe-1, MAI-Voice-1, and MAI-Image-2 — and placed them in Microsoft Foundry. The labels contain no mention of OpenAI. Six months after revising the contract that previously restricted Microsoft from independently developing advanced AI systems, this looks not like a casual experiment but as a declaration of independence.

The context makes this move particularly significant. Since 2019, Microsoft has invested $13 billion in OpenAI — more than any other investor. Azure became the primary infrastructure for training OpenAI's models.

Microsoft itself sold GPT-4 to corporate clients through Azure OpenAI Service. The partnership was so close that the previous contract explicitly prohibited Microsoft from developing its own frontier models — those competing with OpenAI in scope. This ban disappeared six months ago when terms were revised.

And almost immediately afterward, Microsoft began openly demonstrating independent AI development. The three models cover three key areas. MAI-Transcribe-1 specializes in speech recognition and transcription — a task critical for corporate services: Teams, Copilot, office applications.

MAI-Voice-1 handles voice synthesis and generation. MAI-Image-2 works with images. All three are available through Microsoft Foundry — a cloud platform for enterprise AI where organizations deploy, test, and integrate AI models into their own products.

Foundry is now becoming something more than just an Azure showcase for OpenAI. It's a marketplace where Microsoft's models compete directly with OpenAI's models. A corporate client can choose: pay for GPT-4o or use the MAI series.

This fundamentally changes Microsoft's negotiating position — having your own alternative always redistributes the balance of power between partners. This is not about a breakup. Microsoft remains OpenAI's largest investor and continues to integrate their models into Copilot and Azure OpenAI Service.

But Microsoft remembers well how dependence on a single supplier ends. In November 2023, when OpenAI's board unexpectedly fired the then-CEO, Microsoft's head found himself in a situation where the fate of the company's key AI direction was being decided without his involvement. The crisis was resolved, but the lesson was learned.

Since then, Microsoft has systematically built up its own AI capabilities: research divisions, model development teams, computational infrastructure. The MAI series is the visible result of this course. For OpenAI, the situation is becoming difficult.

Their largest distributor and investor is creating direct competition within the same platform. This is not the first time in technology history that an ecosystem absorbs a startup that gave it birth: Amazon began competing with sellers on its own marketplace, Google preinstalls its own applications, sidelining third-party ones. The pattern is familiar.

The industry has long been moving toward this point. Google has Gemini, Amazon is building Nova, Meta releases open-source Llama. Microsoft could not remain the only major tech player without its own model lineup — especially having invested tens of billions of dollars in AI infrastructure.

The MAI series is not the final chapter of the story but a new one. Microsoft is ceasing to be merely a distributor and is becoming a competitor. With its own name on the label.