Jack Parker-Holder (@jparkerholder) / X

Jack Parker-Holder

1,305 posts

Jack Parker-Holder

@jparkerholder

Research Director scaling world models @GoogleDeepMind, Honorary Associate Professor @UCL_DARK. Dad (🧒👶🐶), CFC fan, BJJ. Views are my own :)

London, England

jparkerholder.github.io

Joined October 2018

Pinned
Jack Parker-Holder
@jparkerholder
Aug 5, 2025
Genie 3 feels like a watershed moment for world models 🌐: we can now generate multi-minute, real-time interactive simulations of any imaginable world. This could be the key missing piece for embodied AGI… and it can also create beautiful beaches with my dog, playable real time
00:00
2.1M
Jack Parker-Holder
@jparkerholder
Aug 5, 2025
This one is an example where you see what someone did with your model and can’t even believe it
Tim Rocktäschel
@_rockt
Aug 5, 2025
Replying to @_rockt
Actions are persistent in the world.
00:00
230K
Jack Parker-Holder
@jparkerholder
Dec 4, 2024
Introducing 🧞Genie 2 🧞 - our most capable large-scale foundation world model, which can generate a diverse array of consistent worlds, playable for up to a minute. We believe Genie 2 could unlock the next wave of capabilities for embodied agents 🧠.
00:00
2.6M
Jack Parker-Holder
@jparkerholder
Aug 6, 2025
This is probably one of the most mind blowing things the model can do... and so meta
Jakob Bauer
@jkbr_ai
Aug 6, 2025
Something we discovered by accident: what happens if we start Genie 3 from a video and a completely unrelated prompt? Turns out the model really, really wants to make it work, to the point where it emulates itself. The prompt in this one is about a trex on a tropical island.
00:00
31K
Jack Parker-Holder
@jparkerholder
Jul 25, 2022
I'm super excited to be joining @DeepMind today as a Research Scientist, working with @_rockt! Thank you to everyone who helped make this possible! Watch this space 🌱
Jack Parker-Holder
@jparkerholder
Nov 4, 2025
Excited to build models that simulate the world? 🌐🌐🌐 Our team is hiring!! See posts below for Research Scientists and Engineers. The roles are somewhat fluid, but rule of thumb is the RS typically require a PhD. Preference for London or Toronto. Apply here: 👇👇
150K
Jack Parker-Holder
@jparkerholder
Jan 12, 2022
🤖 Introducing the first survey on AutoRL: methods for automatically discovering multiple components of the RL training pipeline, from tuning hyperparameters and architectures to learning algorithms or automatically designing environments. Link 👉 arxiv.org/abs/2201.03916 [1/4]
Jack Parker-Holder
@jparkerholder
Aug 5, 2025
Replying to @jparkerholder
A year ago we wouldn’t have believed this was possible, but this is the benefit of working with such an incredible group of people @GoogleDeepMind . Shout out to my co-lead @shlomifruchter and the core team who went above and beyond. Excited to get this out to the world soon 🚀
46K
Jack Parker-Holder
@jparkerholder
Mar 3, 2022
Evolving Curricula with Regret-Based Environment Design Website: accelagent.github.io Paper: arxiv.org/abs/2203.01302 TL;DR: We introduce a new open-ended RL algorithm that produces complex levels and a robust agent that can solve them (e.g. below). Highlights ⬇️! [1/N]
00:00
Jack Parker-Holder
@jparkerholder
Dec 7, 2024
🚨🚨 We are hiring a student researcher to start in Q1 2025!! If you're interested in exploring the potential for world models to fuel the next wave of embodied AI, apply here by *DECEMBER 13TH*. google.com/about/careers/…
Jack Parker-Holder
@jparkerholder
Dec 4, 2024
Introducing 🧞Genie 2 🧞 - our most capable large-scale foundation world model, which can generate a diverse array of consistent worlds, playable for up to a minute. We believe Genie 2 could unlock the next wave of capabilities for embodied agents 🧠.
00:00
42K
Jack Parker-Holder
@jparkerholder
Feb 26, 2024
When we started this project the idea of training world models *exclusively* from Internet videos seemed wild, but it turns out latent actions are the key and the bitter lesson holds. Now we have a viable path to generating the rich diversity of environments we need for AGI. 🚀
Tim Rocktäschel
@_rockt
Feb 26, 2024
I am really excited to reveal what @GoogleDeepMind's Open Endedness Team has been up to 🚀. We introduce Genie 🧞, a foundation world model trained exclusively from Internet videos that can generate an endless variety of action-controllable 2D worlds given image prompts.
GIF
49K
Jack Parker-Holder
@jparkerholder
Oct 9, 2025
Super cool to see Genie 3 recognized as one of @TIME's Best Inventions of 2025!! Congrats to the incredible team for making it possible :)
Google DeepMind
@GoogleDeepMind
Oct 9, 2025
We’re proud to announce that Genie 3 has been named one of @TIME’s Best Inventions of 2025. Genie 3 is our groundbreaking world model capable of generating interactive, playable environments from text or image prompts. Find out more → goo.gle/3KGqiYa
15K
Jack Parker-Holder
@jparkerholder
Aug 6, 2025
With Genie 3 we can now step into our existing videos and move around in real time.
Jakob Bauer
@jkbr_ai
Aug 6, 2025
Yesterday we announced Genie 3. One feature of the model that's especially fun to play with is starting worlds from existing videos. Here's a drone shot generated by Veo 3, with me taking control mid-flight.
00:00
16K
Jack Parker-Holder
@jparkerholder
Dec 4, 2024
Replying to @jparkerholder
🤯🤯🤯… And just like that, we have a path to unlimited environments for training and evaluating our embodied agents! We tried creating another world with three arches, and once again Genie 2 was able to simulate the world and SIMA solved the task ✅.
00:00
67K