Why It Accidentally Got Called Nano Banana 🍌 | Made by Google Podcast S8E8

Product & Technology - Google's Gemini app team developed an image generation model initially known as "Nano Banana," officially named "Gemini 2.5% Flash image" [1][30] - The "Nano Banana" model achieved character and facial consistency, allowing users to generate images that closely resemble themselves, loved ones, or pets [10] - The model gained popularity on LM Arena, an external website for ranking different models, and trended on X (formerly Twitter), leading to higher-than-expected traffic [14][15] - Google uses visible watermarks and SynthID (an invisible, unbreakable watermark) to indicate AI-generated images from the Gemini app, addressing concerns about authenticity and responsible use [38][39] - A new model on top of the V03 series, 3.1%, is an improvement across the board on quality, highest in what Google calls photo to video [44] Usage & Trends - Users have created over 5 billion images on the Gemini app since launch [16] - Viral trends included figurines, Polaroid-style images, and restoration of old photos, with users finding both humorous and emotional applications [22][24][26] - The initial viral trend was a figurine, popularized by celebrities and other users, originating in Thailand with 90-word prompts [22][23] - Photo to video has been brought to the EU and UK for the first time [45] Future Development - Google is actively gathering user feedback (thumbs up/down, complaints) to inform future improvements and launches for the Gemini app [42][43]