OpenAI will introduce “thinking” AI “Strawberry” in 2 weeks

OpenAI will introduce a new AI called “Strawberry” as part of its popular ChatGPT service. Its launch is expected within the next two weeks, approximately on September 26. A distinctive feature of Strawberry will be the ability to “think through” responses before issuing them, which can significantly improve the quality of user interaction with the AI ​​chatbot.

Image source: Mariia Shalabaieva / unsplash.com

This data is based on feedback from two people who have already tested the new AI model. As a pioneer in the development of large language models, OpenAI continues to drive not only interest in AI, but also significant investment in the field. All this is confirmed by the growing number of companies seeking to integrate AI into their products and services. OpenAI, which receives significant support from Microsoft, recently reported that it has more than a million users with paid subscriptions. This impressive growth in the user base is largely due to the widespread adoption of the ChatGPT chatbot.

«Strawberry” stands out among other AIs with its unique approach to processing user requests. While most existing AI models strive to respond quickly, Strawberry is able to “think through” its answers before delivering them. This innovative “reflection” process has the potential to lead to deeper, more accurate, and contextually relevant responses, significantly improving the user experience.

It is expected that the new AI will be integrated into ChatGPT, but will remain a separate product. However, the exact format for providing it to users remains unclear. It is important to note that the first version of Strawberry will be limited in its capabilities – it will only be able to process and generate text information. This means that at this stage the system is not multimodal and is not capable of working with images.