Google debuts Google AI Edge Foresight, a macOS note-taking app for capturing meetings with Google's local 740M-parameter EmbeddingGemma 2 and Gemma 4 assistant

Local, offline AI assistants for meetings are now available on macOS, without needing an internet connection.

Google has launched Google AI Edge Foresight, a new macOS note-taking application designed for capturing meeting information. This app operates as a local-first competitor to tools like Granola and functions entirely offline. It leverages Google's on-device EmbeddingGemma 2 model, which has 740 million parameters, to transcribe and summarize meetings, including in-person events. The application features a split-screen interface allowing users to take manual notes alongside AI-generated summaries. Users can also access a full meeting transcript or interact with a Gemma 4-powered assistant for specific queries. Additionally, the app supports building a knowledge base by uploading various document types such as PDFs, Google Docs, and Microsoft Office files, enabling real-time answers based on this information. Google AI Edge Foresight is optimized for Apple Silicon, ensuring its offline capabilities are fully realized.

Google's introduction of AI Edge Foresight signals a growing trend towards on-device AI processing for sensitive data like meeting notes. By relying on local models such as EmbeddingGemma 2 and Gemma 4, Google is demonstrating a commitment to user privacy and offline functionality, potentially challenging cloud-dependent solutions. This move could push competitors to accelerate their own offline AI development or enhance the security features of their cloud-based offerings.

The competitive landscape for AI meeting assistants is intensifying with Google's entry, joining players like Granola, Wispr, Calendly, and Superhuman's Fathom. AI Edge Foresight's integration of a local knowledge base and conversational AI capabilities sets a new benchmark for feature sets. The emphasis on offline operation, particularly for Apple Silicon users, suggests a strategic focus on specific hardware ecosystems and a potential pathway for future cloud-integrated Gemini applications.