Microsoft's GenAI Team Pioneers Compact Language Models for Cost-Efficient AI Solutions

Reducing Computing Power and Financial Barriers with Smaller Language Models

NEWS  AI  January 24, 2024  Reading time: 2 Minute(s)

Max (RS editor)

---

In a strategic move to harness the potential of artificial intelligence (AI) with reduced computing power and financial overheads, Microsoft is expanding its focus on developing Smaller Language Models (SLMs). This initiative aims to empower Microsoft's AI products and services with conversational and generative capabilities similar to large language models (LLMs) such as OpenAI's GPT, while substantially decreasing the associated hardware requirements.

According to a recent report from The Information, Microsoft has established the GenAI team, a dedicated unit anticipated to be integrated into the Azure cloud platform. The team's primary mission is to craft compact language models that rival or surpass the functionality of their larger counterparts, while prioritizing efficiency in terms of size, cost, and environmental impact.

To fortify the GenAI team, Microsoft has enlisted top developers from its research group, with corporate vice president Misha Bilenko leading the charge. The objective is clear: develop AI solutions that offer advanced capabilities at a fraction of the traditional computational and financial costs associated with large-scale language models.

Microsoft's incorporation of AI capabilities from OpenAI into its existing products, such as Bing utilizing GPT-4 for Copilot and Bing Image Creator leveraging Dall-E for image generation, demonstrates a history of collaboration. However, the recent emphasis on building in-house expertise suggests a strategic shift toward reducing dependence on external providers.

The GenAI initiative aligns with Microsoft's broader goals of making AI more accessible and sustainable. By creating smaller language models, Microsoft aims to minimize carbon footprints and costs associated with AI technologies. This not only benefits the company's bottom line but also opens up new possibilities for delivering customized AI solutions directly to consumers.

 

In light of OpenAI's recent GPT Store launch, enabling users to create, share, and sell customized AI models, Microsoft has announced its plans to offer custom AI chatbots tailored for Copilot users. This move is strategic, offering users more control over their AI solutions and reducing reliance on external providers.

 COVER IMAGE BY KJPARGETER ON FREEPIK / REVIEW SPACE