Google has announced a new AI model, Genie 4, “which unlocks unprecedented realism and improves real-time interactive simulation.”

Google has been playing catchup to Anthropic and OpenAI, with its Gemini models often lagging behind in capability. Until recently, Google’s models were largely competing on price and efficiency. Recently, however, Google has made major strides toward closing the gap with its Gemini 4 Argon model, launched in September.

The company may have closed the gap altogether with Genie 4, its new world model. The company says Genie 4 is “natively multimodal, enabling audio-like music, ambient noises, and interactive dialogue to be simultaneously generated with the world.”

In the race to AGI (artificial general intelligence), understanding how the world works is a critical step in that process, and one that traditional LLMs have struggled with. Genie 4 is designed to further advance that understanding.

Understanding and simulating the world is a crucial aspect of building AGI.…