当AI视频从128像素进化到以假乱真，它掌握的已不仅是画面生成，更是真实的物理规律。
本期视频由 Runway 联合创始人兼 CEO Cristóbal Valenzuela 深度分享，揭秘下一代视频生成模型的演进路径。通过海量数据的“下一帧预测”，模型正在潜移默化地掌握重力、遮挡与物体恒存性等物理常识。
- 从生成到模拟：视频模型正演变为实时交互的“世界模型”，允许用户在生成环境中直接干预并观察物理反馈。
- 机器人学与界面融合：世界接口模型（World Interface Models）将赋予机器人更直观的环境理解力。
- Continuum 原型：探索将屏幕上的任何元素转化为实时、可交互的视频流，重塑未来的操作系统与视觉编程体验。
这不仅是视频生成的迭代，更是通向全模拟世界的必经之路。
----------
原视频标题：You Can't Tell What's Real Anymore. Here's What Comes Next - Cristóbal Valenzuela, Runway
原视频作者：AI Engineer
原视频链接：https://www.youtube.com/watch?v=q4ITwvlFplQ
原视频发布时间：2026-11-03 08:00:29
原视频简介：Runway's first AI film in 2019 was 128x128 pixels. Today, people can't tell generated video from real footage. Runway co-founder and CEO Cristóbal Valenzuela explains how next-frame prediction at scale teaches models physics, object permanence and how the world works. He then shows what happens when those models become real-time and interactive: world interface models, robotics, and Continuum, an early research prototype where anything on your screen can become a live, interactive video.
0:00 Intro
0:50 Runway's first AI film (2019)
1:49 State-of-the-art video today
2:48 Video as next-frame prediction
4:04 What video models learn: physics and object permanence
5:54 From video to simulation
6:54 Making world models real-time and interactive
7:59 Robotics and world interface models
10:04 Extrapolating forward: Continuum
13:45 Visual thinking for coding agents
15:05 Live, reactive experiences
16:08 Everything will be simulated
