TechFlow News, April 2: According to JINSHI Data, Microsoft (MSFT.O) plans to develop a large, state-of-the-art AI model by next year to build an internal alternative to OpenAI’s and Anthropic’s most advanced AI tools. Mustafa Suleyman, CEO of Microsoft AI, stated in an interview: “We must deliver technology that is truly at the absolute forefront. By 2027, our goal is for our models to achieve genuine state-of-the-art performance in text, image, and audio generation and response.” On Thursday, Microsoft’s AI division launched a speech-to-text transcription model, reportedly outperforming competitors on benchmark tests across 11 of the 25 most widely used languages. However, like the division’s previously released speech and image generation models, this model is an optimized, specialized tool trained on significantly less data than general-purpose models such as Claude 3 Opus or OpenAI’s GPT-4. Suleyman noted that Microsoft is consolidating computing resources to build broader-capability models, having begun expanding its computational infrastructure with a set of NVIDIA GB200 chips last October. He added: “Over the next 12 to 18 months, we will progressively scale up to frontier-level computing capacity.”
Risk Disclosure: This website's content is not investment advice and offers no trading guidance or related services. Per regulations from the PBOC and other authorities, users must be aware of virtual currency risks. Contact us / support@techflowpost.com ICP License: 琼ICP备2022009338号
