Elon Musk has shared a ‘throwback’ post from 2016 that captured a key moment in the history of AI: Nvidia CEO Jensen Huang personally delivering the company’s first AI-optimised GPU server, the DGX-1, to OpenAI. At the time, OpenAI was a non-profit organisation co-founded by Musk and Sam Altman. However, apart from being a critical moment in the tech world, Musk has used the post to promote new feature in Grok AI chatbot that allows people to convert a photo into a video.“Jensen delivering the first AI-optimized GPU to OpenAI in 2016,” said Musk in the post on X in which the Tesla CEO shared a 6-second video of Jensen Huang unpacking a GPU. However, upon closer inspection, the video appears to be AI-generated.The original photo shows Huang handing over the DGX-1. Valued at $129,000, the supercomputer was touted by Nvidia as the “world’s first supercomputer dedicated to artificial intelligence.” It was built with eight interconnected GPUs and could perform in two hours what would take traditional CPUs six days to complete.Musk has made multiple subsequent posts related to Grok’s new feature, further strengthening the fact that the video is not authentic but an AI-version of the original photo.
Elon Musk’s xAI launches ‘Make video with Grok’ feature
Musk’s AI company, xAI, recently launched a new feature called “Grok Imagine” that allows users to create videos from text or images. The tool, which is an extension of the Grok chatbot, is currently available to users on the Grok iOS app and is being rolled out to Android users as well.Grok Imagine allows users to generate short videos with sound from a text prompt or from a still image that they provide. It includes four style options for video generation: Normal, Fun, Custom, and Spicy.“Long tap on any image and turn it into a video in ~17 seconds,” Musk said in a post on X.“And the @Grok app can turn any images on your phone into a video. This is by far the biggest use for most people. Show friends how to do it!” he added.