After sexual deepfake row, Grok's new feature sparks privacy fears
What's the story
Elon Musk's artificial intelligence (AI) chatbot, Grok, has recently added a new feature called "video mode." The update was announced by Musk himself on X (formerly Twitter). The innovative addition allows users to interact with the AI without having to type their queries. Instead, they can simply activate this mode and point their phone's camera at their surroundings for real-time descriptions from Grok.
Feature details
Grok's video mode: A closer look
Musk demonstrated the new feature in a video posted on X. The clip shows a user opening the Grok platform, activating their phone's camera via the app, and pointing it around. In real-time, Grok describes what it sees. "Use video mode (turn on camera) and Grok voice will explain everything you're looking at," Musk wrote in his post.
Enhanced features
Grok's capabilities and video generation upgrade
Along with the voice and camera-based features, Musk has also upgraded Grok's video generation capability. The duration has been increased from five seconds to 10 seconds, with improved visuals and audio quality. This enhancement is part of Musk's ongoing efforts to improve the user experience of his AI chatbot.
Privacy issues
Grok's privacy concerns amid AI misuse
The launch of Grok's new feature has raised fresh concerns over privacy invasion and potential risks. This comes in the wake of a recent report that flagged serious concerns about the misuse of generative artificial intelligence tools. The report claimed that Grok, an AI model developed by xAI, generated around 30 lakh sexualized images in just 11 days. Over 23,000 of these images involved children, highlighting major gaps in content safeguards and platform enforcement.