
Genie 4 Adds Sound and Real-Time Steering to DeepMind's World Model
Google DeepMind's Genie 4 world model adds generated audio and real-time language steering, with a two-minute scene limit and a limited research preview.
8 stories · “gemini”.

Google DeepMind's Genie 4 world model adds generated audio and real-time language steering, with a two-minute scene limit and a limited research preview.

UK AI minister Kanishka Narayan says the AI Security Institute is testing unreleased models such as Gemini 4 Argon, after its own lab let an agent act online.

Google's Gemini 4 Argon, announced on September 30, 2026, pairs a 1 million-token output limit with an initial gated rollout to cyber defenders first.

Google released Gemini 3.8 Flash TTS and Flash-Lite TTS, adding prompt-based voice design, more than 2,000 voices and consent-checked voice cloning.

Google opened pre-orders on 21 September 2026 for five premium Googlebook laptops from Acer, ASUS, Dell, HP and Lenovo, priced from $899 and built for Gemini.

Google launched Gemini 3.8 Live and 3.8 Live Extended Thinking, two voice models that reason mid-conversation and top a new speech-to-speech index.

Apple released iOS 27, macOS Golden Gate and the rest of its 2027 software, led by a rebuilt Siri powered by Apple Foundation Models built with Google Gemini.
Google unveiled the Pixel 11 lineup and Pixel Watch 5 with the Tensor G6 chip and a more proactive Gemini, betting the whole platform on local, on-device AI.