Six months after renegotiating the contract that once prevented it from independently pursuing frontier AI, Microsoft has launched three internal models that directly challenge the partner it spent $13 billion to cultivate. MAI-Transcribe-1, MAI-Voice-1, and MAI-Image-2 are now available from Microsoft Foundry and do not carry the OpenAI name anywhere on the label.
The models are the first publicly released result from the MAI Superintelligence team that Microsoft AI CEO Mustafa Suleyman formed in November 2025 with the stated mission of pursuing what the company calls “humanistic superintelligence.” In an internal memo from March 1st reported by Business InsiderSuleyman wrote that he intended to focus all his energy on superintelligence and deliver world-class models for Microsoft over the next five years. That ambition now has its first tangible evidence.
MAI-Transcribe-1 is, on paper, the most disruptive of the three. The speech-to-text model claims to have the lowest word error rate across 25 languages in the FLEURS benchmark, averaging 3.8 percent, and Microsoft says it beats OpenAI’s Whisper-large-v3 in all 25 languages, Google’s Gemini 3.1 Flash in 22 out of 25, and ElevenLabs’ Scribe v2 in 15 out of 25. It runs 2.5 times faster than the previous transcription service Azure Fast from Microsoft and is priced at $0.36 per hour of audio. Perhaps most telling is the team that built it: just 10 people.
MAI-Voice-1 completes the audio loop. The text-to-speech model generates 60 seconds of natural-sounding audio in less than a second on a single GPU and supports the creation of custom speech from a few seconds of sample audio. Combined with MAI-Transcribe-1 and a large language model chosen by the customer, it forms a complete voice channel that runs entirely on Microsoft infrastructure without any dependency on OpenAI technology.
MAI-Image-2, the largest of the three, had already debuted at number three on Arena.ai’s text-to-image rankings in March, placing it behind only Google’s Gemini 3.1 Flash and OpenAI’s GPT Image 1.5. The model was developed in collaboration with photographers, designers and visual storytellers, and WPP, one of the world’s largest marketing groups, is among the first business partners to build it at scale.
Strategic context matters more than reference points. Until the September 2025 renegotiation, Microsoft’s original partnership agreement with OpenAI contractually prevented the company from independently pursuing overall AI development. The revised memorandum of understanding fundamentally changed that calculus. Microsoft retained licensing rights to everything OpenAI builds through 2032, secured $250 billion in new commercial commitments in the Azure cloud, and, crucially, gained the freedom to build competitive models. Suleyman acknowledged the twist directly: The contract renegotiation, he said, allowed Microsoft to independently exercise its own superintelligence.
The timing is deliberate. Jacob Andreou, former senior vice president at Snap, took over as executive vice president of Copilot on March 17, freeing Suleyman from daily product responsibilities. The MAI models arrived just two weeks later. Microsoft also hired Ali Farhadi, former executive director of the Allen Institute for AI, to Suleyman’s superintelligence team in March, a recruiting sign that ambitions extend far beyond transcription and imaging.
For OpenAI, development creates an uncomfortable dynamic. Microsoft remains its largest investor and largest cloud infrastructure provider, and the two companies continue to share a platform in Foundry, which hosts the OpenAI and Microsoft models. But OpenAI’s own push towards commercial monetization is accelerating in parallel, and the relationship is starting to look like two companies orbiting the same market with overlapping products rather than a partnership with a clear division of labor. OpenAI’s $110 billion raise in February, backed by SoftBank, Nvidia and Amazon, valued the company independently of Microsoft at a level that makes the original partnership framework increasingly anachronistic.
The broader market for AI models is fragmenting along similar lines. Anthropic raises $30 billion at $380 billion valuation established it as a credible third force in enterprise AI, with revenues of $14 billion. Google continues to rapidly iterate on Gemini. The era when OpenAI was the only game in town for cutting-edge AI capabilities, and Microsoft was content to be its exclusive distribution channel, is definitely over.
Microsoft Foundry, the platform formerly known as Azure AI Foundry and before that Azure AI Studio (the second rebrand in twelve months), now serves developers at more than 80,000 companies, including 80 percent of Fortune 500 companies. That distribution advantage is what makes the MAI family of models strategically significant: Microsoft doesn’t need to outperform OpenAI in every benchmark to shift enterprise spending toward internal models. It must be competitive enough that customers choose the integrated option over the third-party alternative, a dynamic that the last year of AI industry consolidation has become increasingly plausible.
Suleyman has said it will be another year or two before the superintelligence team produces cutting-edge language models. What landed this week is the foundation: a multimodal toolset that gives Microsoft its own voice, ears, and eyes independent of OpenAI. The $13 billion partnership has no end. But the premise it was built on, that Microsoft needed OpenAI to compete in AI, is being It quietly dismantled one model release at a time..
