The AI Podcast Runway Raises $315M for AI World Models
Feb 11, 2026
A deep dive into Runway's $315M funding and its push to build next-gen AI world models. Conversation covers how 3D simulation and physics-aware systems improve video generation. Discussion of Gen 4.5 capabilities like HD text-to-video and native audio. Examination of responsible watermarking, consent policies, and emerging deepfake regulations shaping the space.
AI Snips
Chapters
Transcript
Episode notes
World Models Solve Video Consistency
- World models create internal 3D simulations so video AI can reason about space, physics, and time.
- These simulations let the model render consistent camera motion and object behavior across frames.
Prebuilt 3D Scenes Enable Creative Camera Moves
- Runway uses 3D simulated environments so users can freely move the camera without glitches.
- Prebuilding the scene enables smooth panning, rotations, and creative camera effects.
Runway's Early Physics-Aware Success
- Runway first gained traction with a physics-aware video generation tool used in media and advertising.
- They recently released their first world model and now see it as foundational for many industries.