Google just dropped its Nano Banana AI image generator into your search results and notes (www.techradar.com)

Google has quietly embedded its Nano Banana image generator (from the Gemini 2.5 Flash family) directly into Search via Lens/AI Mode and into NotebookLM’s Video Overviews. In Search, a new Create tab lets you snap, upload or select photos and apply text prompts to transform them—or generate images from scratch. In NotebookLM, Nano Banana powers context-aware visuals for narrated Video Overviews, offering six stylistic treatments (Watercolor, Papercraft, Anime, Whiteboard, Retro Print, Heritage) and two output lengths (deep-dive “Explainer” and short “Brief”). Visuals are pulled from your documents so animations illustrate the actual content rather than generic stock art. For the AI/ML community this signals continued product-level deployment of generative vision models: Google is moving beyond standalone demos to bake image synthesis into everyday workflows (Search, notes, soon Photos). Nano Banana is touted as high-volume and consistent—used to produce over five billion images and designed to preserve facial detail, object layout and style across outputs—though Google concedes occasional prompt misinterpretations. The rollout underscores trends toward context-aware multimodal generation, tighter UX integration (compare Microsoft Copilot, Adobe and OpenAI efforts), and raises practical tradeoffs between speed, polish and occasional errors as these models become default tooling for students, creators and enterprise users.