Following the release of Grok 2 in August, which introduced image generation capabilities to the AI chatbot, Musk-owned xAI has released another update, which has given the chatbot the ability to understand and interpret images and answer questions about them.
Paid X (formerly Twitter) users can upload an image into Grok, and it will tell them what it is and where it’s from, and give them extra context about the content within the image, when prompted. Musk even claims it can understand humor, so could potentially even interpret funny memes and GIFs.
Musk has established that although it’s still in its “early stages” it will “rapidly improve”, and will be able to interpret other file types and documents, like PDFs, for example, soon. He’s also boasted, in typical Musk fashion, that although rivals—like OpenAI’s ChatGPT, Anthropic’s Claude, and Google’s Gemini—already have this image understanding capability (and have had for a while), “We are getting done in months what took everyone else years.”