OpenAI has introduced a neural network called CLIP that can learn visual concepts from natural language supervision. This means it can understand what objects or scenes are described in text, such as 'a photo of a cat' or 'a cityscape'. CLIP can be applied to various visual classification tasks by simply providing the names of the categories to be recognized. --- Why it matters: This matters because CLIP demonstrates a significant step forward in bridging the gap between natural language understanding and computer vision. It has implications for applications such as image captioning, object detection, and visual question answering. Source: https://openai.com/index/clip