In an exciting development, Meta Platforms announced on Tuesday its release of an advanced artificial intelligence model known as I-JEPA. This cutting-edge model, described as "human-like," offers researchers access to its components and boasts superior capabilities in analyzing and completing unfinished images when compared to existing models.
Unlike other generative AI models that rely solely on nearby pixels, I-JEPA utilizes background knowledge about the world to fill in missing elements of images. By incorporating human-like reasoning, as championed by Meta's esteemed AI scientist Yann LeCun, this innovative approach helps the technology avoid common errors found in AI-generated images, such as hands with extra fingers.
Meta Platforms, the parent company of Facebook and Instagram, has consistently demonstrated its commitment to open-sourced AI research through its in-house research lab. CEO Mark Zuckerberg believes that sharing the models developed by Meta's researchers not only fosters innovation but also aids in identifying safety gaps and reducing costs.
Zuckerberg expressed, "We find it highly advantageous if the industry adopts the same fundamental tools we employ. This way, we can benefit from the improvements made by others."
Despite concerns raised by industry peers regarding the potential risks associated with this technology, Meta's executives chose not to endorse a statement last month supported by top executives from OpenAI, DeepMind, Microsoft, and Google. This statement equated the risks of AI with those posed by pandemics and wars.
LeCun, often regarded as one of the "godfathers of AI," has passionately refuted "AI doomerism" and advocated for the integration of safety checks into AI systems.
Furthermore, Meta is actively integrating generative AI features into its consumer products. This includes advertising tools capable of generating image backgrounds and an Instagram feature that modifies user photos based on text prompts.
Meta unveils USD499 Quest 3 virtual, mixed reality headset