Microsoft Unveils Three Proprietary AI Models to Challenge OpenAI and Google
Key Points
- Microsoft launches three proprietary AI models: MAI‑Transcribe‑1, MAI‑Voice‑1, MAI‑Image‑2.
- Models are available through the Microsoft Foundry platform and MAI Playground.
- MAI‑Transcribe‑1 supports 25 languages and runs 2.5× faster than Azure Fast.
- MAI‑Voice‑1 creates 60 seconds of natural‑sounding audio in one second and allows custom voice creation.
- MAI‑Image‑2 ranks in the top three on the Arena.ai image‑generation leaderboard.
- Rollouts underway in Bing (image search) and PowerPoint (graphic generation).
- Pricing undercuts comparable services from Amazon and Google.
- Release follows the expiration of a 2019 OpenAI contract that barred Microsoft from building its own frontier AI.
- Mustafa Suleyman affirms continued partnership with OpenAI despite the new models.
Microsoft announced the launch of three in‑house AI models—MAI‑Transcribe‑1, MAI‑Voice‑1 and MAI‑Image‑2—through its Foundry platform and MAI Playground. The models, designed for speech‑to‑text, synthetic voice and image generation, aim to rival offerings from OpenAI, Google and Amazon. Built after a 2019 contract with OpenAI lifted a restriction on Microsoft’s own frontier AI work, the new suite promises faster performance, multilingual support and competitive pricing, with rollouts already underway in Bing and PowerPoint.
Microsoft rolled out a trio of proprietary artificial‑intelligence models on Tuesday, positioning the company to compete directly with OpenAI, Google and Amazon in core generative‑AI services. The models—MAI‑Transcribe‑1, MAI‑Voice‑1 and MAI‑Image‑2—are now accessible via the Microsoft Foundry platform and a public MAI Playground, giving developers a ready‑to‑use toolkit for speech, voice and visual tasks.
MAI‑Transcribe‑1 targets speech‑to‑text conversion across 25 languages and delivers results 2.5 times faster than Azure’s Fast offering. A lean team of ten engineers built the model, emphasizing both speed and accuracy. MAI‑Voice‑1 generates natural, expressive audio at a rate of 60 seconds of speech in just one second of compute. The model also supports custom voice creation from a brief audio sample, opening new possibilities for personalized assistants and media production. MAI‑Image‑2, the newest addition to Microsoft’s image‑generation portfolio, has secured a top‑three spot on the Arena.ai leaderboard, signaling strong performance against rival systems.
Rollouts have already begun in consumer‑facing products. Bing now taps MAI‑Image‑2 for enhanced visual search, while PowerPoint integrates the model to generate graphics on demand. The company’s broader AI strategy, long anchored by its partnership with OpenAI, is shifting to incorporate these home‑grown models into the backbone of services like Copilot and Teams, where they have quietly powered features behind the scenes.
The timing of the release follows a contractual change that lifted a restriction on Microsoft’s ability to develop frontier AI. A 2019 agreement with OpenAI granted Microsoft cloud credits and model licensing but barred the firm from building its own competing models. That clause expired in October 2025, freeing Microsoft to pursue its own research agenda. CEO of Microsoft AI, Mustafa Suleyman, reiterated the firm’s commitment to the OpenAI partnership even as the new models signal a parallel path.
Pricing for the MAI suite undercuts comparable offerings from Amazon and Google, a move that could attract cost‑conscious developers and enterprises. Analysts note that the models’ performance and price point may make them the preferred foundation for Microsoft’s expanding AI product portfolio, especially as the company seeks to reduce reliance on external providers.
Industry observers will watch how quickly developers adopt the models and whether the integration into flagship products drives measurable gains in user engagement. For now, Microsoft’s AI push marks a decisive step toward a more self‑sufficient ecosystem, challenging the dominance of established generative‑AI players while leveraging its deep cloud infrastructure and enterprise reach.