Meta, the tech giant behind Facebook, Instagram, WhatsApp, and Threads, is making waves in the AI world with two major announcements that showcase its rapid advancements.
While many still associate AI with companies like Google and OpenAI, Meta is quietly pushing boundaries with its latest innovations. The company has just unveiled two exciting AI features: a cutting-edge model that can help robots think and act intuitively, and a new AI-powered video editing tool aimed at millions of users.
First up is V-JEPA 2, a revolutionary AI model that is designed to give robots and AI agents the ability to understand and predict the world around them, much like humans do. Imagine walking through a busy area, dodging obstacles and anticipating what’s next without having to consciously think about it. This is exactly what Meta aims to teach robots with V-JEPA 2 – the ability to “think ahead” by building a world model.
V-JEPA 2 helps AI in three main ways: it enables understanding, prediction, and planning. The model is trained on video footage, allowing it to observe and learn how objects and people interact in real-life scenarios. This new version of V-JEPA has improved on the original model from last year, making it more adept at handling new environments and tasks that robots may encounter for the first time.
During testing, Meta’s robots demonstrated impressive abilities, such as reaching out to grab objects and moving them around. While this may seem basic, it’s a significant achievement in the field of robotics, as it showcases the potential of AI to perform tasks that were once thought to be difficult for machines. However, Meta is not the only player in this space. Google’s Gemini 2.0 and OpenAI’s AI agents also aim to enhance task management and reasoning capabilities. Still, Meta’s focus seems to be on developing tools that offer tangible benefits to users, which could reshape how we interact with AI in the future.
Moving on to the second announcement, Meta has rolled out an AI-powered video editing tool that allows users to remix short-form videos in creative ways. This feature, available on the Meta AI app, Meta.AI website, and the new Edits app, lets users transform their videos using preset prompts that can change everything from the background to the overall style of the clip.
The tool offers over 50 editing prompts to choose from, letting users get creative by turning their videos into comic book scenes, adding dreamy sparkles to cloudy shots, or creating neon-lit video game aesthetics. While currently free for editing up to 10 seconds of video, this feature is just the beginning. Meta plans to roll out text-based editing prompts later this year, giving users even more control over their video content.
Once users are happy with their edits, they can easily share their creations on Facebook, Instagram, or the Meta AI app. Although this feature is fun and innovative, it also signals a larger trend in the industry. As AI-driven editing tools become more sophisticated, the line between reality and digital manipulation may continue to blur. With tools like Meta’s video editor and Google’s hyper-realistic video generator, the potential for both creativity and deception is endless.