Article complete

Google DeepMind Unveils Genie 3 as Step Toward AGI With Real-Time 3D World Generation

Access to Genie 3 is currently limited to a small group of academic researchers and creative professionals. 

Google DeepMind Unveils Genie 3 as Step Toward AGI With Real-Time 3D World Generation

Key takeaways

  • Unveil Genie 3, a groundbreaking AI model generating real-time interactive 3D environments.
  • Enhance user experience with seamless navigation and object retention over extended interactions.
  • Introduce promptable world events, allowing dynamic changes through simple text commands.
  • Position Genie 3 as a crucial step toward achieving artificial general intelligence (AGI).
  • Limit access to Genie 3 for select academic researchers and creative professionals only.

Takeaways drafted with AI from the article text.

“Given a text prompt, Genie 3 can generate dynamic worlds that you can navigate in real time at 24 frames per second, retaining consistency for a few minutes at a resolution of 720p,” the company said in its blog post. Previous models, such as Genie 2 allowed only around 10–20 seconds of interaction.

Unlike earlier versions, Genie 3 retains objects in place even when users move the camera away and return later. This ability to remember elements such as wall markings or object positions is being presented as a step forward in the development of AI-driven virtual environments.

Genie 3 also introduces “promptable world events,” allowing users to change weather or add characters using simple text commands. This brings a level of control that may prove useful for researchers and creators experimenting with dynamic environments.

“World models are also a key stepping stone on the path to AGI, since they make it possible to train AI agents in an unlimited curriculum of rich simulation environments,” Google DeepMind said.

Access to Genie 3 is currently limited to a small group of academic researchers and creative professionals. DeepMind says this closed release will help assess potential safety issues and explore new ways of interacting with AI-generated worlds.

Genie 3 builds on previous iterations by improving resolution, frame rate, interaction time and environment stability. The release is positioned as part of DeepMind’s broader efforts in building general-purpose AI systems capable of planning and reasoning in complex simulated settings.

“Genie 3 could create new opportunities for education and training, helping students learn and experts gain experience. Not only can it provide a vast space to train agents like robots and autonomous systems, Genie 3 can also make it possible to evaluate agents’ performance and explore their weaknesses,” the company said.

There is no public release timeline for Genie 3. DeepMind says wider availability may be considered after further research and safety testing.

Published

Corrections: AIM corrects errors of fact and notes the change on the story. Spotted one? Tell the editors. Read our editorial standards.

More in this story

Google Deepmind · 5 stories
  1. NVIDIA’s KeyNote, Microsoft’s Build Conference And More: Top AI News This Week
  2. Google DeepMind Launches Lyria, Transforming the Future of Music with AI
  3. Google DeepMind's Table Tennis Robot Competes at Human-Level
  4. You are here: Google DeepMind Unveils Genie 3 as Step Toward AGI With Real-Time 3D World Generation
  5. Google DeepMind Unveils SIMA 2, an AI Agent That Acts and Learns in 3D Environments
About the author

Tech Journalist

Siddharth is a media graduate who loves to explore tech through journalism and putting forward ideas worth pondering about in the era of artificial intelligence.

All stories
Belamy · Daily

The day’s AI news, before your first meeting.