The biggest gap between what video/world models promise and what they deliver is controllability — creators want to fine-tune every nuance like knobs in traditional software, and current models can't yet decompose to that level of granular control, on top of imperfect physics simulation.
Asked about failure modes, Zeve says the core gap isn't just physics accuracy but controllability — creators want knob-level control over every detail, which current models can't yet offer. ✦ AI generated
Zeve Farman · The Cognitive Revolution · 2026-07-08 · original ↗
starts at this moment · 79:50
“Zeve, one area that often gets less airtime is failure modes in the creative pipeline. Where do you see the biggest gap between what your tools can reliably deliver today and what creators assume they'll get? And how are you testing against that?”
Most creators are still going to point at the fact that uh the simulation isn't as correct as as controllable as they want it to be, right? Like if you are talking to really creative people, they typically want to control like every nuance of the appearance and that uh requires to somehow decompose the model to like a bunch of knobs
verbatim transcript · starts at 79:50
79:50like 200 or 500 billion parameters the entire physics of the universe of at least a par universe as we know we are not there yet but we're closing the gaps pretty quickly like I would say probably the Most creators are still going to point at the fact that uh the simulation isn't as correct as as controllable as they want it to be, right? Like if you are talking
80:16to really creative people, they typically want to control like every nuance of the appearance and that uh requires to somehow decompose the model to like a bunch of knobs, right? Like that you have like in the classical software. we can say yeah here I want to have to more light and here I want like the splash to be bigger so achieving this controllability right so besides like
80:38the physics is also like one of the I would say open things right like we we're getting cool things not necessarily the things that creators wants exactly and it is a pain point thanks Q thanks um anything else that you think is important that we haven't touched on and I appreciate you've uh stayed with us a long. I don't want to uh take up your whole afternoon into evening here, but
81:05um this has been a great conversation so far. Anything that you feel like we we and Q have missed? Notice like a great question. I feel like the edge compute angle is underpriced. People should kind of spend more time thinking about that. And listen, your intuition about like when is the right time during the training process to fuse modalities. That's like something that uh again like for me personally is a cool question
81:34where we spend a time of time. So I have to say frankly you we discussed like a ton of cool and important stuff. >> Cool. Well, I love it. Um, thanks for joining us today on AI in the AM. And uh, we'll definitely be watching your progress and continuing to make music videos with it. And, uh, you know, who knows what we might be able to do,