Theoretically Media: AI Video Just Stepped Into the Real World
The channel's uploads, logged episode by episode: 198K subscribers, 22,036,803 views and 420 videos. Nothing here is summarised from memory — every entry carries the episode's own outline and a view count read on 2026-11-01.

Key takeaways
- Theoretically Media has 198K subscribers and 22,036,803 views across 420 videos, read 2026-11-01.
- 10 of the 15 uploads in this log are long form; the rest are Shorts.
- The most-viewed episode in the window is AI Video Just Stepped Into the Real World! at 76,885 views.
- Every episode below links to the video and carries the channel's own outline of its contents.
Updated 1 Nov 2026 · next update 8 Nov 2026
Stills: Theoretically Media.
Key facts as of 2026-11-01
- @TheoreticallyMedia has 198K subscribers and 22,036,803 views across 420 videos — the channel page, read 2026-11-01.
- He publishes most days: of the last 15 uploads, 10 run long form (over 90 seconds) and 5 are Shorts. — the feed
- The most-viewed episode in the current window is AI Video Just Stepped Into the Real World!, at 76,885 views — watch it, read 2026-11-01.
- 4 episodes landed in the week to 1 Nov 2026: AI Video Just Stepped Into the Real World!; Genie 4 Is the FUTURE of AI World Models!; Kling IPO Signals a "Voltron" AI Video Model?; Free & Open Source AI Video Previz Tool!.
The channel is run by one person in a room, one take at a time, and it is the fastest honest read on AI video that exists: he tests what shipped this week rather than what was announced, and he says when a tool is not worth the money.
What changed this week
2026-10-29 · AI Video Just Stepped Into the Real World! — 14:21, 76,885 views, read 2026-11-01.
Your AI characters can now stand in your actual room, and react to what's in it!
Simulon is back: turn a character image into a 3D "neural asset," place it in the real world through your phone's camera, act out the scene with your own camera move, and the app regenerates the take with the character responding to what you do and say.
Also: a kitbash into Kling 4.0 and Genie 4, and Alibaba's Apsara slide with Wan,…
2026-10-27 · Genie 4 Is the FUTURE of AI World Models! — 13:46, 61,266 views, read 2026-11-01.
Google DeepMind just announced Genie 4, and I've had early access for a few days. It's a real-time world model: describe a place, drop into it, and change it while you're inside, now with sound and characters who talk. With Gemini 4 Argon planning in the background, you can basically vibe-code your own interactive worlds.
LINKS
Genie 4 announcement (Google DeepMind):…
2026-10-26 · Kling IPO Signals a "Voltron" AI Video Model? — 0:40, 3,770 views, read 2026-11-01.
A little "Inside Baseball" on the Chinese AI Video models, and while I don't normally cover International Business News, I thought this was pretty interesting.
The Voltron of AI Video models, next year?
2026-10-25 · Free & Open Source AI Video Previz Tool! — 0:56, 6,314 views, read 2026-11-01.
CozyClay 3.0: a free, open-source 3D previs tool built for AI video. One-click install through Pinokio; it runs in your browser and doesn't need a GPU.
Full details in this video:
https://youtu.be/qlSI_ZbRdEc
The episode log
Every episode in the current window, newest first, with its own view count read on 2026-11-01. Long-form episodes carry the channel's own outline of what is in them.
AI Video Just Stepped Into the Real World!
2026-10-29 · Long form · 14:21 · 76,885 views · 1,787 likes (read 2026-11-01)
Your AI characters can now stand in your actual room, and react to what's in it! Simulon is back: turn a character image into a 3D "neural asset," place it in the real world through your phone's camera, act out the scene with your own camera move, and the app regenerates the take with the character responding to what you do and say. Also: a kitbash into Kling 4.0 and Genie 4, and Alibaba's Apsara slide with Wan, Happy Horse and Happy Shrimp on the board. WHAT'S IN THIS ONE
- Simulon, back from 2023 (2.0 shipped…
Genie 4 Is the FUTURE of AI World Models!
2026-10-27 · Long form · 13:46 · 61,266 views · 1,360 likes (read 2026-11-01)
Google DeepMind just announced Genie 4, and I've had early access for a few days. It's a real-time world model: describe a place, drop into it, and change it while you're inside, now with sound and characters who talk. With Gemini 4 Argon planning in the background, you can basically vibe-code your own interactive worlds. LINKS Genie 4 announcement (Google DeepMind): https://deepmind.google/blog/genie-4-a-new-generation-of-world-simulation/ Genie 4 waitlist (Project Genie): https://forms.gle/WQnSNEM3omF7Uruu7 My…
Kling IPO Signals a "Voltron" AI Video Model?
2026-10-26 · Short · 0:40 · 3,770 views · 47 likes (read 2026-11-01)
A little "Inside Baseball" on the Chinese AI Video models, and while I don't normally cover International Business News, I thought this was pretty interesting. The Voltron of AI Video models, next year?
Free & Open Source AI Video Previz Tool!
2026-10-25 · Short · 0:56 · 6,314 views · 162 likes (read 2026-11-01)
CozyClay 3.0: a free, open-source 3D previs tool built for AI video. One-click install through Pinokio; it runs in your browser and doesn't need a GPU. Full details in this video: https://youtu.be/qlSI_ZbRdEc
New Mystery AI Video Model: Iron Goddess!
2026-10-24 · Short · 0:56 · 2,099 views · 31 likes (read 2026-11-01)
- Iron Goddess: an unclaimed AI video model that showed up in battle mode on the Arena this week. Big, fast, swing-for-the-fences motion; frame by frame, the fastest moments get smudgy. Seedance 2.5 holds together better, with less dynamic movement.
- Who made it? Nobody has said. My bet is Alibaba, which said at its Apsara conference in September that a new video model is coming in November.
This FREE AI Previz Tool is Awesome & a Secret Video Model!
2026-10-22 · Long form · 17:38 · 53,461 views · 1,220 likes (read 2026-11-01)
A free, open-source tool that gets you the exact shot before you generate, plus the fix for when 3D-guided AI video comes out stiff. And a secret AI video model just turned up in blind tests on the Arena, swinging big on motion. Also this week: Kling's reported IPO, Ideogram's first vector model, and a Seedream 5.5 leak. WHAT'S IN THIS ONE
- Iron Goddess: an unclaimed AI video model that showed up in battle mode on the Arena this week. Tests from Synthesis (fluid dynamics), Leo (a character shot), Brent Lynch…
FREE 3d City Generator!
2026-10-19 · Short · 0:52 · 7,247 views · 220 likes (read 2026-11-01)
World Gacha: a free browser tool by Sharp (@sharp_maker) that generates fictional towns from random numbers and an erosion sim, with street level and 3D export. Free for commercial use. Worth a bookmark or just something fun to play with for five minutes. OR BOTH! World Gacha: https://tools.oyasumi-gamedev.com/worldgacha/
ByteDance's New World Model Can Re-Shoot Scenes!
2026-10-17 · Short · 0:52 · 4,845 views · 102 likes (read 2026-11-01)
ByteDance's unreleased world model took a scene from Stephen Chow's King of Comedy and re-shot it from new camera angles off one drawn camera path, at a recent showcase event. More details on the World Model in this video: https://youtu.be/VTrc346ea30
ByteDance's World Model Re-Shoots Movies, Sora Returns? & Krea 3 Is Here!
2026-10-15 · Long form · 16:33 · 54,041 views · 1,403 likes (read 2026-11-01)
ByteDance's new world model, Sora returns?, Krea 3 is here, OmniHuman 2.0 is back, Pico Space Pro, World Gacha, and LTX-2.5's new post tools. ByteDance's unreleased world model took a scene from Stephen Chow's King of Comedy and re-shot it from new camera angles off one drawn camera path, at Sunday's Douyin gala. Then: OmniHuman 2.0 (API only, not in the US yet), Krea 3's beta, a very fast AI image editor, the report that Sora is coming back via a Jeffrey Katzenberg and Bill Peebles startup, and a free browser…
Google Cooked! Testing Gemini 4 Argon & Blasting 800 Million Tokens!
2026-10-13 · Long form · 15:14 · 50,172 views · 1,071 likes (read 2026-11-01)
Google is finally out of "Flash land." Gemini 4 Argon is out, and Iâve been running it in early access as my daily driver for the past week. Even better: Google put it on the $20 Pro plan (and didn't introduce a $500 tier). Today weâre looking at what I actually built with it â including a turn-based Godot & Blender game where the AI betrays humanity, a one-shot polyphonic synth, an overnight unattended Skyrim modding spree using /goal mode, cutting a 90-minute interview down to 10 minutes in DaVinci Resolve…
What No One Tells You About Making a 17-Minute AI Backrooms Film
2026-10-08 · Long form · 16:15 · 26,989 views · 670 likes (read 2026-11-01)
A 17-minute Backrooms short film, made by one person with AI. Javi Lopez, co-founder of Magnific, walks me through how he pulled it off, the "don't do handheld" lesson, and the thing no one tells you about making a film this long alone: form a band. This is the first of what might become a recurring thing: other AI filmmakers on the channel, so I can steal their workflows and pass them on to you. The longest AI short I've made is about eight minutes, so I wanted to know how Javi handled a 17-minute monster, and…
Make Any Game With AI & 2 New Open Source Video Models!
2026-10-06 · Long form · 15:15 · 72,182 views · 1,331 likes (read 2026-11-01)
AI coding agents can now mod almost any PC game you own, so I put DOOM inside Skyrim. SkyDoom is real Skyrim with all the NPCs, except Doom monsters spawn everywhere and I have a chaingun. Built with UniversalModder in about an hour, and I'll show you how to mash up any two games you own. Also this week: two new open source video models in one day (Prism does native 2K with audio but wants an 80 GB card; Kandinsky 6.0 runs on a 4090), Nano Banana 2.1 is out and it's a 0.1, and Hark wants to be your AI best…
Kling 4! Gemini 4! ElevenLabs v4! Ideogram 4.5! Everything Goes Four!
2026-10-01 · Long form · 13:36 · 52,941 views · 1,266 likes (read 2026-11-01)
Four is your lucky number today. Kling 4 is here â kinda. But, I ran it through the usual tests: text-to-video, image-to-video, Omni with Flamethrower Girl, a noir scene in Malloy's office, and my keyframes-plus-Omni trick. Then Google wakes up with Gemini 4 Argon (announced, not released â return of the King? We'll see), HeyGen drops a video model that starts at a penny a second (asterisk included), Ideogram 4.5 gets precise editing, and ElevenLabs ships Eleven v4. AI fall is getting toasty. WHAT'S IN THIS…
OpenAI Cooked with Dots, But What Will It Cost You?
2026-09-29 · Long form · 13:37 · 64,602 views · 916 likes (read 2026-11-01)
OpenAI's DevDay was all about dots, ChatGPT's new always-on AI agents â basically its catch-up flex on Meta's Muse. So did OpenAI cook? And if so, how much is that meal going to cost? On the menu: ChatGPT dots (on Pro from $100/mo, rolling out now), ChatGPT Space, GPT-6.1 Sol and some new pricing that may or may not make everyone happy. My dot hasn't shown up yet (I'm always the last one), but I have spent a few days with Meta's Muse, the AI agent that knocked ChatGPT off No. 1 on the US App Store â and I have…
Meta Muse Won? Alibaba's "Seedance Killer" & AI Audio Levels Up!
2026-09-24 · Long form · 17:04 · 71,737 views · 1,208 likes (read 2026-11-01)
A blockbuster week for AI â and not just for the frontier labs. Anthropic and OpenAI dropped Opus 5.5 and GPT-6 Sol (and Luna) on the same day, Meta brought Muse storming into hardware at Meta Connect â including the Muse Charm, which is basically an AI Tamagotchi with its own 5G â and OpenAI has Dev Day next week with the Aeon rumors swirling. The battleground is pretty clear: everyone wants to be your AI best friend. Then, on our side of the fence: Alibaba teased a November video model that, in bootleg…
Why we cover him
We added this page because the same man keeps turning up in our own reporting as the first honest test of whatever launched that week, and because his episode notes are the closest thing this industry has to a primary document: what he actually ran, on what, and what it did when he ran it.
Sources
- The channel's upload feed: https://www.youtube.com/feeds/videos.xml?channel_id=UC9Ryt3XOGYBoAJVsBHNGDzA
- The channel page, for subscribers, total views and video count: https://www.youtube.com/@TheoreticallyMedia/about
- AI Video Just Stepped Into the Real World! (2026-10-29): https://www.youtube.com/watch?v=mpqgn9DMld4
- Genie 4 Is the FUTURE of AI World Models! (2026-10-27): https://www.youtube.com/watch?v=kJgYSiIZMb8
- Kling IPO Signals a "Voltron" AI Video Model? (2026-10-26): https://www.youtube.com/watch?v=QwzgUvqpleg
- Free & Open Source AI Video Previz Tool! (2026-10-25): https://www.youtube.com/watch?v=rUpGzeoXwTs
- New Mystery AI Video Model: Iron Goddess! (2026-10-24): https://www.youtube.com/watch?v=U6vf0kjMPqU
Sources
- youtube.com - tier 1 — the channel's own upload feed
- youtube.com - tier 1 — subscribers, total views, video count
- youtube.com - tier 1 — the episode itself, 2026-10-29
- youtube.com - tier 1 — the episode itself, 2026-10-27
- youtube.com - tier 1 — the episode itself, 2026-10-26
- youtube.com - tier 1 — the episode itself, 2026-10-25
- youtube.com - tier 1 — the episode itself, 2026-10-24