Microsoft has announced plans to develop a cutting-edge AI model by 2027, aiming to rival the most advanced tools from OpenAI and Anthropic. Mustafa Suleyman, Microsoft's CEO of AI, emphasized the company's commitment to achieving state-of-the-art capabilities in text, image, and audio generation. This initiative follows the recent launch of a speech transcription model that reportedly outperformed competitors in 11 benchmarks across 25 languages. Microsoft is consolidating its computing power to enhance its AI capabilities, having expanded its computational resources with Nvidia GB200 chips since last October. Suleyman stated that the company plans to scale up to frontier-level computing capabilities over the next 12 to 18 months.