Google Launches Offline AI Notetaking App for Mac Using New EmbeddingGemma 2 Model

Google’s new Edge Foresight app uses the EmbeddingGemma 2 model to convert handwritten scribbles into structured notes entirely on-device.

Threat Times Systems DeskOctober 9, 20262 min read

Share LinkedIn Reddit

Illustration: A sleek silver laptop sits on a wooden desk with its screen displaying a digital notebook interface filled with clean, abstract geometric shapes. — AI-generated illustration: Threat Times

Key points

Google announced the EmbeddingGemma 2 model for on-device information organization and search.
The Edge Foresight Mac app demonstrates the model by converting offline handwritten notes.
The system processes data locally on consumer hardware without requiring cloud connectivity.

Google has announced the release of a new artificial intelligence application for macOS called Edge Foresight, which leverages the company’s newly introduced EmbeddingGemma 2 model. According to 9to5Mac, the app is designed to transform handwritten scribbles into full, structured notes using offline capabilities.

The primary function of Edge Foresight is to serve as an intelligent notetaking tool that operates entirely on the user's hardware. By utilizing the EmbeddingGemma 2 model, the application can organize, search, and connect information without transmitting data to external servers. This on-device approach highlights Google’s strategy to integrate advanced AI features directly into consumer electronics, ensuring that sensitive user data remains local. The announcement positions the new model as a tool for enhancing productivity through immediate, private data processing.

Model Capabilities and On-Device Processing

The EmbeddingGemma 2 model is the underlying technology powering the Edge Foresight application. According to 9to5Mac, this model is specifically engineered to organize, search, and connect information directly on consumer hardware. This capability allows for real-time processing of handwritten input, converting informal scribbles into coherent digital notes.

The emphasis on "offline" functionality indicates that the AI inference happens locally on the Mac’s processor. This design choice addresses privacy concerns associated with cloud-based AI services, as no user data needs to leave the device to be processed. The model’s ability to connect disparate pieces of information suggests advanced contextual understanding, allowing users to retrieve specific details from their notes through natural language search or automated organization.

Impact on Mac Users and Developers

The release of Edge Foresight targets Mac users who rely on digital or scanned handwritten notes for work or personal organization. By enabling offline AI processing, Google addresses a niche for users who require high privacy standards or who work in environments with limited internet connectivity. The application serves as a proof-of-concept for what is possible with the EmbeddingGemma 2 model, potentially encouraging other developers to build similar on-device AI tools for macOS.

For security professionals, the shift toward on-device AI processing changes the threat landscape. While data privacy is improved by keeping information local, it also means that security controls must be robust at the endpoint level. If the device is compromised, the locally stored AI models and processed data could be at risk, shifting the focus from network security to device integrity and access control.

What should you do?

Security teams should monitor the deployment of new on-device AI models to understand their resource requirements and potential attack surfaces. Users should verify that the application has proper permissions and that their system’s firewall and endpoint protection are up to date. Organizations should assess whether on-device AI tools align with their data retention and privacy policies.

Background: Privacy policies

A privacy policy is a legal document that explains what data a company collects, why it collects it, and how it shares that information. It defines your rights regarding consent and deletion. Understanding it helps you control your digital footprint and avoid unintended data exposure to third parties.

Read the full guide: How Privacy Policies Actually Work: A Practical Explanation

What to do and how to stay safe: Google

Verify that your macOS system is updated to the latest version to ensure compatibility and security patches for new AI applications.
Review the permissions requested by the Edge Foresight app to ensure it only accesses necessary data, such as notes and handwriting inputs.
Monitor system performance and resource usage, as on-device AI processing can be computationally intensive and may impact battery life or system responsiveness.
Check for any vendor advisories regarding the EmbeddingGemma 2 model to stay informed about potential vulnerabilities or updates.

General security guidance from the Threat Times newsroom. It is not confirmed advice from the organisations named in this story.