Microsoft Copilot soon will be able to generate responses using the OpenAI GPT-4 Turbo model, which will enable users to tackle more complex and longer-running tasks. Credit: Thinkstock After launching the unified Microsoft Copilot and Copilot Studio two weeks ago, Microsoft is now preparing more advanced capabilities for Copilot that will be available early next year. Setting sights on 2024, Microsoft on December 5 said Copilot soon will be able to generate responses using the OpenAI GPT-4 Turbo model, which will enable users to tackle more complex and longer-running tasks. The GPT-4 Turbo model is being tested with select users and is due to be integrated into Microsoft Copilot in coming weeks. Also, the company said users already can use Copilot to create higher-quality images, and images more accurate to the prompt, using an updated DALL-E 3 model. These capabilities are accessible now via bing.com/create or by prompting Copilot to create an image. Additionally, Microsoft is developing a code interpreter that will perform complex tasks such as more accurate calculations, data analysis, coding, visualization, and math. The company is gathering feedback on these capabilities from select users, and plans to make the code interpreter widely available soon. Microsoft also is combining the GPT-4 with Vision model with Bing image search and web search data to offer better image understanding for queries. Finally, Microsoft said Edge browser users soon will be able to use Copilot to rewrite text from most websites. Users need only select the text to change and ask Copilot to rewrite it. Microsoft recently made Copilot accessible to anyone on any device. The company in mid-November also launched Copilot Studio, a low-code development tool for customizing Microsoft Copilot for Microsoft 365 and for building AI assistants. Generative AIArtificial IntelligenceTechnology Industry