← November 20, 2026 briefing · added in the 12:00 KST update

Genie 4 follow-up: Google opens its world model to developers via the Gemini API (Interactions API) and AI Studio (report)

Announced November 19, 2026

Verified (fact)

According to AI Times Korea (인공지능신문), Google announced Genie 4 on the 18th (local time) and on the 19th opened developer access through the Gemini API's Interactions API and Google AI Studio. Developers can add world-model features to their own apps and services. Genie 4 reportedly generates 720p virtual environments from text and reference images, which users can explore in real time and control with natural language. The article does not give pricing or usage limits.

Why it matters (interpretation)

World models have moved past the demo stage into a product developers can call through an API. They can be used for games and simulation, as well as for generating robot training environments.

Confidence medium · official source pending

Sources

Story thread

  1. 2026-11-18 Google DeepMind announces that its world model Genie 4 is now available
  2. 2026-11-20 Genie 4 follow-up: Google opens its world model to developers via the Gemini API (Interactions API) and AI Studio (report)

More about Google