AI Weekly Malaysia

Google's AI Edge team released Google AI Edge Foresight, a Mac app positioned against the AI note-taker Granola, which runs entirely offline using an on-device EmbeddingGemma 2 model with 740 million parameters and a Gemma 4-powered assistant for querying notes. It offers a Granola-style split screen (manual shorthand on one side, AI notes on the other), full transcripts, and a knowledge base built from uploaded PDFs, Google Docs, Microsoft Office formats, plain text, Markdown, and web bookmarks, and it can surface relevant answers in real time during a meeting. Google says it is optimized for Apple Silicon and works with no internet connection; the article notes this may be as much a showcase for the Gemma offline model series as a mainstream product, and no pricing or non-Mac availability is given.

Why: If you build or buy meeting/note tooling, this is a concrete data point that a 740M-parameter on-device embedding model plus Gemma 4 is enough for transcription-plus-retrieval on a laptop with no cloud calls — meaning no per-seat transcription bills and no meeting audio leaving the machine. The catch is the Apple Silicon-only constraint: if your team or customers are on Windows or Linux, this specific app is not a path, so treat it as evidence the local stack works rather than a tool you can standardize on today. No pricing or general availability was announced, so don't plan a migration off Granola or cloud note-takers on the strength of this alone.

Microsoft opened preorders for the Surface Laptop Ultra, a $2,599 machine containing an Nvidia chip that the company says can run large AI models locally without generating cloud bills; it ships Oct. 16. Microsoft calls it the world's first laptop with a magnetic USB-C port (MagSafe-style) alongside two standard USB-C ports, announced by Pavan Davuluri, EVP of Windows and devices, at a San Francisco event. Microsoft also said Meta will bring its Muse AI agent to Windows.

Why: The pitch is capital cost instead of metered inference: pay $2,599 (roughly RM11,000) once and run models on-device rather than paying per-token API bills. If you are budgeting local inference for agent or coding workloads, the decision is whether that one-off hardware price beats your projected API spend over the machine's life — and the article does not name the chip, the models, or performance figures, so that math cannot be done from this text alone. The Meta Muse agent for Windows is the more immediately relevant item for anyone building agent tooling on Windows, but the announcement gives no date, pricing, or capability detail.

Sigil Wen, described in the article as a self-taught coder and Thiel Fellow, launched an invite-only beta of Underdog on Monday — a fully on-device AI assistant that currently runs on Macs and Windows PCs, with Linux, iPhone, and Android versions stated as coming soon. It uses a custom inference engine called Husky that Wen says moves less data between the main chip and the graphics chip than other on-device engines, encrypts the keys to accounts like email that users authorise, and runs a 27-billion-parameter reasoning model fine-tuned from an unnamed base — considerably smaller than data-centre state-of-the-art models. The article reports no benchmarks, latency figures, model licence, pricing, or beta size.

Why: The concrete trade-off is the 27B parameter ceiling plus no published benchmarks — so if you are considering on-device inference for privacy-sensitive workloads (for example, anything you would not want leaving a Malaysian-hosted app under PDPA obligations), you cannot yet compare Underdog against a cloud agent on quality or speed. The other practical blocker is platform: invite-only, Mac and Windows only, with Linux, iOS, and Android not yet shipped, so Linux-first or mobile-first teams cannot test it at all right now. Treat the 'most private' framing as an unverified claim until the encrypted-key design and Husky engine are independently examined.