• Generative AI's Impact on Tech Giants: The rapid expansion of generative AI has changed Google and other tech giants' product design. Most AI features use remote servers with high processing power, while phones have less power.
• Google's New Mobile AI Tools: At I/O next week, Google will likely announce new APIs for on-device AI using Gemini Nano. It has quietly published documentation for developers, with an update to the ML Kit SDK adding API support.
• Gemini Nano's Features and Limitations: Gemini Nano is built on AI Core and similar to the experimental Edge AI SDK. It can enable apps to do summarization, proofreading, etc. without sending data to the cloud, but has limitations like a maximum of three bullet points in summaries and only English image descriptions. Different versions like Gemini Nano XS and XXS have different sizes and context windows.
• Benefit to Android: This move is good for Android as ML Kit works on devices outside Google's Pixel line. Other phones like OnePlus 13, Samsung Galaxy S25, and Xiaomi 15 are also designed to run this model.
• Importance for App Developers: App developers are in a tough spot without proper tools. Google's new APIs make implementing local AI easier. On-device models are more secure as they parse information locally instead of sending it to remote servers.
• Need for Collaboration: The release of APIs relying on Google and OEMs' collaboration on Gemini Nano may face challenges as some companies might go their own way and not all phones have enough power to run AI locally.