正在加载视频...

视频加载失败

返回首页

I tested Google’s world model Genie 3... Then DeepMind told me everything 00:00 - Intro & Authoring Workflow 00:27 - Genie 3 Playtesting & Demos 05:33 - Interview w/ Google DeepMind (Genie 3 co-lead Jack Parker-Holder and Sr. PM Diego Rivas) 06:54 - Wildest emergent behaviors observed 09:02 -...

bilawalsidhu's profile picture

Bilawal Sidhu

107,935 subscribers

13,243 次观看 • 4 个月前 •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Why make explorable, AI generated worlds? Shlomi Fruchter and Jack Parker-Holder explain how creating diverse and challenging virtual environments can help test and train AI agents safely. 🦺 Watch their conversation about Genie 3 with podcast host, Hannah Fry ↓ Timecodes: 00:00 Coming up 00:54 Intro 02:00 What is Genie 3? 03:00 Demos 06:50 Physics 11:00 Possible scenarios 15:07 Emergent properties 20:35 Veo and Genie 25:00 Hunter's lodge 28:02 Memory 34:52 SIMA and Genie 3 39:13 Searching for interestingness 52:00 AGI and the future 59:23 Hannah's thoughts
1:00:40

Why make explorable, AI generated worlds? Shlomi Fruchter and Jack Parker-Holder explain how creating diverse and challenging virtual environments can help test and train AI agents safely. 🦺 Watch their conversation about Genie 3 with podcast host, Hannah Fry ↓ Timecodes: 00:00 Coming up 00:54 Intro 02:00 What is Genie 3? 03:00 Demos 06:50 Physics 11:00 Possible scenarios 15:07 Emergent properties 20:35 Veo and Genie 25:00 Hunter's lodge 28:02 Memory 34:52 SIMA and Genie 3 39:13 Searching for interestingness 52:00 AGI and the future 59:23 Hannah's thoughts

Google DeepMind

184,644 次观看 • 10 个月前

Today, we're joined by Jack Parker-Holder and Shlomi Fruchter, researchers at Google DeepMind, to discuss the recent release of Genie 3, a model capable of generating “playable” virtual worlds. We dig into the evolution of the Genie project and review the current model’s scaled-up capabilities, including creating real-time, interactive, and high-resolution environments. Jack and Shlomi share their perspectives on what defines a world model, the model's architecture, and key technical challenges and breakthroughs, including Genie 3’s visual memory and ability to handle “promptable world events.” Jack, Shlomi, and Sam share their favorite Genie 3 demos, and discuss its potential as a dynamic training environment for embodied AI agents. Finally, we will explore future directions for Genie research. 🗒️ For the full list of resources for this episode, visit the show notes page: 📖 CHAPTERS =============================== 00:00 - Introduction 7:11 - What is a world model? 14:49 - Milestones of Genie research 24:32 - Genie 3 27:46 - Challenges 30:07 - Genie 3 examples 33:48 - Model capabilities 35:49 - Key aspects of the model 39:40 - Consistency as an emergent property 42:11 - Promptable word events 47:24 - SIMA agent 50:56 - Limitations 56:08 - Future directions
1:00:31

Today, we're joined by Jack Parker-Holder and Shlomi Fruchter, researchers at Google DeepMind, to discuss the recent release of Genie 3, a model capable of generating “playable” virtual worlds. We dig into the evolution of the Genie project and review the current model’s scaled-up capabilities, including creating real-time, interactive, and high-resolution environments. Jack and Shlomi share their perspectives on what defines a world model, the model's architecture, and key technical challenges and breakthroughs, including Genie 3’s visual memory and ability to handle “promptable world events.” Jack, Shlomi, and Sam share their favorite Genie 3 demos, and discuss its potential as a dynamic training environment for embodied AI agents. Finally, we will explore future directions for Genie research. 🗒️ For the full list of resources for this episode, visit the show notes page: 📖 CHAPTERS =============================== 00:00 - Introduction 7:11 - What is a world model? 14:49 - Milestones of Genie research 24:32 - Genie 3 27:46 - Challenges 30:07 - Genie 3 examples 33:48 - Model capabilities 35:49 - Key aspects of the model 39:40 - Consistency as an emergent property 42:11 - Promptable word events 47:24 - SIMA agent 50:56 - Limitations 56:08 - Future directions

The TWIML AI Podcast

12,043 次观看 • 10 个月前

My conversation with the leads of the Google Genie 3 team. I asked about simulation theory :) 0:00 Intro 1:04 Genie's Purpose 4:40 Genie + Agents 10:12 More than Video Games? 15:19 Genie's Capabilities 17:23 Genie's Attention to Detail 19:46 Veo 3 & Nano Banana 22:49 Google's Custom Hardware 24:10 Training/Testing Genie 27:09 Benchmarks 30:40 Simulation Theory 38:41 Prompted Events 44:47 Training Data 46:09 Text vs Video Models 49:16 Scale of Genie 50:17 General Availability
52:18

My conversation with the leads of the Google Genie 3 team. I asked about simulation theory :) 0:00 Intro 1:04 Genie's Purpose 4:40 Genie + Agents 10:12 More than Video Games? 15:19 Genie's Capabilities 17:23 Genie's Attention to Detail 19:46 Veo 3 & Nano Banana 22:49 Google's Custom Hardware 24:10 Training/Testing Genie 27:09 Benchmarks 30:40 Simulation Theory 38:41 Prompted Events 44:47 Training Data 46:09 Text vs Video Models 49:16 Scale of Genie 50:17 General Availability

Matthew Berman

77,534 次观看 • 9 个月前

Oh wow.. Nano Banana Pro + Genie 3 is literally a time machine 🤯 Generate image in Nano Banana Pro then render in Genie 3 with the same prompt. Music by Suno AI PROMPT: "31.7785° N, 35.2296° E, April 3, 33 AD, 15:00 hours."
0:59

Oh wow.. Nano Banana Pro + Genie 3 is literally a time machine 🤯 Generate image in Nano Banana Pro then render in Genie 3 with the same prompt. Music by Suno AI PROMPT: "31.7785° N, 35.2296° E, April 3, 33 AD, 15:00 hours."

Min Choi

527,294 次观看 • 4 个月前

The biggest thing I’m looking forward to from Deepmind is Genie 4 Genie 1 – February 23, 2024 Genie 2 – December 4, 2024 Genie 3 – August 5 2025 Genie 1 → Genie 2: 285 days Genie 2 → Genie 3: 244 days So if we stay on trend maybe we will get Genie 4 – somewhere around April–June 2026, I hope for the former month Exciting stuff Demis Hassabis !
0:33

The biggest thing I’m looking forward to from Deepmind is Genie 4 Genie 1 – February 23, 2024 Genie 2 – December 4, 2024 Genie 3 – August 5 2025 Genie 1 → Genie 2: 285 days Genie 2 → Genie 3: 244 days So if we stay on trend maybe we will get Genie 4 – somewhere around April–June 2026, I hope for the former month Exciting stuff Demis Hassabis !

Chris

24,226 次观看 • 6 个月前

I got the Google DeepMind team to test Genie 3 with one of my artworks
1:06

I got the Google DeepMind team to test Genie 3 with one of my artworks

Rui Huang

904,103 次观看 • 10 个月前

Genie 3 from Google DeepMind is the most mind-blowing technology I've ever seen since starting MLST. This is our exclusive in-person interview with Jack Parker-Holder and Shlomi Fruchter
58:22

Genie 3 from Google DeepMind is the most mind-blowing technology I've ever seen since starting MLST. This is our exclusive in-person interview with Jack Parker-Holder and Shlomi Fruchter

Machine Learning Street Talk

29,719 次观看 • 10 个月前

Today we are announcing Genie 3, a general purpose world model by Google DeepMind that can generate dynamic, interactive environments with a single text prompt. World models are AI that understand facets of the world (like Veo's knowledge of intuitive physics or Genie's mastery of new environments), and serve as a key stepping stone on the path to AGI. Genie 3 is our first world model to allow interaction in real-time, while also improving consistency and realism compared to Genie 2. Learn more ➡️
0:33

Today we are announcing Genie 3, a general purpose world model by Google DeepMind that can generate dynamic, interactive environments with a single text prompt. World models are AI that understand facets of the world (like Veo's knowledge of intuitive physics or Genie's mastery of new environments), and serve as a key stepping stone on the path to AGI. Genie 3 is our first world model to allow interaction in real-time, while also improving consistency and realism compared to Genie 2. Learn more ➡️

Google AI

102,524 次观看 • 10 个月前

Google just turned Street View into a video game. The mother lode of ground level data -- 280 billion real world panoramas, now playable in real time. Here's everything you need to know in 7 mins: 00:00 Genie 3 Grounded In Reality 00:44 Real-time Demos! 03:48 The Bigger Picture 05:36 Current Limitations 06:54 World Model Resources
7:40

Google just turned Street View into a video game. The mother lode of ground level data -- 280 billion real world panoramas, now playable in real time. Here's everything you need to know in 7 mins: 00:00 Genie 3 Grounded In Reality 00:44 Real-time Demos! 03:48 The Bigger Picture 05:36 Current Limitations 06:54 World Model Resources

Bilawal Sidhu

17,423 次观看 • 1 个月前

I got hooked up with Genie 3 (thanks Google DeepMind) and asked it to show me what the future looks like after OpenClaw🦞 has taken over.
1:02

I got hooked up with Genie 3 (thanks Google DeepMind) and asked it to show me what the future looks like after OpenClaw🦞 has taken over.

Lenny Rachitsky

31,885 次观看 • 4 个月前

We finally get to play with Genie 3 and it's actually better than I expected! Check it out! Thanks Google DeepMind for letting me play!
5:57

We finally get to play with Genie 3 and it's actually better than I expected! Check it out! Thanks Google DeepMind for letting me play!

Matt Wolfe

17,125 次观看 • 4 个月前

Genie 3 for when your Veo clip ends too soon. Imagen -> Veo -> Genie 3.
0:30

Genie 3 for when your Veo clip ends too soon. Imagen -> Veo -> Genie 3.

Matt McGill

232,863 次观看 • 10 个月前

Google just took the lead... Gemini 3 + Nano Banana Pro have captured the worlds attention. Let's talk about: > what these models do > best ways to use both models > how other people are using them TIME STAMPS 00:00 Introduction 01:05 Deep Dive into Gemini 3 01:50 Real World Examples of Gemini 3 04:33 Gemini 3 in Google AI Studio 06:33 Building a Landing Page with Gemini 3 09:17 Using Gemini 3 in Cursor 10:47 Interactive Water Cycle Simulation 12:30 Exploring Google Gemini's Capabilities 13:53 Introduction to Nano Banana 15:37 Creating Infographics with Nano Banana 19:08 Nano banana in Krea Nodes 23:41 Future of AI Models
25:14

Google just took the lead... Gemini 3 + Nano Banana Pro have captured the worlds attention. Let's talk about: > what these models do > best ways to use both models > how other people are using them TIME STAMPS 00:00 Introduction 01:05 Deep Dive into Gemini 3 01:50 Real World Examples of Gemini 3 04:33 Gemini 3 in Google AI Studio 06:33 Building a Landing Page with Gemini 3 09:17 Using Gemini 3 in Cursor 10:47 Interactive Water Cycle Simulation 12:30 Exploring Google Gemini's Capabilities 13:53 Introduction to Nano Banana 15:37 Creating Infographics with Nano Banana 19:08 Nano banana in Krea Nodes 23:41 Future of AI Models

Riley Brown

33,706 次观看 • 6 个月前

.Google DeepMind CEO Demis Hassabis joins Logan Kilpatrick on Release Notes to chat about the momentum of AI development, world models like Genie 3, improved model evaluations through Game Arena with @Kaggle, and the quest for AGI.
0:56

.Google DeepMind CEO Demis Hassabis joins Logan Kilpatrick on Release Notes to chat about the momentum of AI development, world models like Genie 3, improved model evaluations through Game Arena with @Kaggle, and the quest for AGI.

Google AI Developers

20,468 次观看 • 10 个月前

Project Genie is a Google Labs experiment that lets you simulate dynamic worlds you can navigate in real time with Genie, our general-purpose world model. Today, we’re connecting Project Genie to nearly 20 years of Street View data from Google Maps — so you can now build interactive spaces based on real-world locations. Street View imagery in Project Genie is available now for places in the U.S., and will expand to more locales over time. #GoogleIO
0:26

Project Genie is a Google Labs experiment that lets you simulate dynamic worlds you can navigate in real time with Genie, our general-purpose world model. Today, we’re connecting Project Genie to nearly 20 years of Street View data from Google Maps — so you can now build interactive spaces based on real-world locations. Street View imagery in Project Genie is available now for places in the U.S., and will expand to more locales over time. #GoogleIO

Google

175,447 次观看 • 1 个月前

🇳🇱🗽 This is Google's Genie 3 World Model with my New Amsterdam (current day New York City) Simulator I created the first frame with Nano Banana Pro on Photo AI and then Min Choi ran it because I can't get access outside US yet Genie 3 lets you control the ship and sail wherever you want, fully interactive, for up to 60 seconds (for now) It also doesn't have sound yet, so I added it myself below, hope you like 😊
0:59

🇳🇱🗽 This is Google's Genie 3 World Model with my New Amsterdam (current day New York City) Simulator I created the first frame with Nano Banana Pro on Photo AI and then Min Choi ran it because I can't get access outside US yet Genie 3 lets you control the ship and sail wherever you want, fully interactive, for up to 60 seconds (for now) It also doesn't have sound yet, so I added it myself below, hope you like 😊

@levelsio

1,124,297 次观看 • 4 个月前

World Models vs Game Engines -- Who Wins? Gaming stocks had a bloodbath after Google's Genie 3. Unity down. Roblox down. Nintendo down. Unity down 35%. Unity's CEO posted a whole thread. Tim Sweeney weighed in. They're both saying the same thing -- world models won't replace game engines, they'll merge together. I tested Genie 3 extensively, broke down the CEO arguments, and mapped out what's actually real versus hype. Including why this might play out exactly like self-driving cars did -- and what that means for the next 3-5 years. 0:00 - Gaming stocks tank after Genie 3 1:12 - Unity CEO's damage control thread 2:51 - Tim Sweeney's response 3:46 - The hybrid approach (and why it has merit) 5:38 - The gap nobody's talking about 10:46 - Why Genie 3 is an entirely new medium 14:38 - Where I actually net out
17:43

World Models vs Game Engines -- Who Wins? Gaming stocks had a bloodbath after Google's Genie 3. Unity down. Roblox down. Nintendo down. Unity down 35%. Unity's CEO posted a whole thread. Tim Sweeney weighed in. They're both saying the same thing -- world models won't replace game engines, they'll merge together. I tested Genie 3 extensively, broke down the CEO arguments, and mapped out what's actually real versus hype. Including why this might play out exactly like self-driving cars did -- and what that means for the next 3-5 years. 0:00 - Gaming stocks tank after Genie 3 1:12 - Unity CEO's damage control thread 2:51 - Tim Sweeney's response 3:46 - The hybrid approach (and why it has merit) 5:38 - The gap nobody's talking about 10:46 - Why Genie 3 is an entirely new medium 14:38 - Where I actually net out

Bilawal Sidhu

18,820 次观看 • 4 个月前

Demis Hassabis describes a surreal setup: one AI playing inside the mind of another DeepMind SIMA agent can now operate within Genie 3 -- taking actions in a world that Genie generates in real time "an AI creates the world, another acts inside it" this setup could unlock unlimited training data for robotics and AGI
1:05

Demis Hassabis describes a surreal setup: one AI playing inside the mind of another DeepMind SIMA agent can now operate within Genie 3 -- taking actions in a world that Genie generates in real time "an AI creates the world, another acts inside it" this setup could unlock unlimited training data for robotics and AGI

Haider.

215,914 次观看 • 10 个月前

Google Genie 3 just did the impossible, recursion inside recursion, a world within a world
0:26

Google Genie 3 just did the impossible, recursion inside recursion, a world within a world

FearBuck

112,297 次观看 • 10 个月前

Demis Hassabis: AGI will be an intertwined model of Gemini, Genie and something like Nano Banana. Nano Banana shows first climps of AGI "It also has Gemini under the hood, it sort of understands what's going on semantically in those images, so I think that's sort of, it's getting towards a kind of AGIfor imaging" AGI will be the unification of the Genie 3, Nano Banana and Gemini models "And then eventually, we got to kind of converge all of those different projects. And they're intertwined. But we need to, you know, converge them all into one" Google seems to be doing everything right at the moment. With Genie 3, Gemini 3, and Nano Banana, they have all the prerequisites to create a general model that develops a true understanding of the real world.
1:45

Demis Hassabis: AGI will be an intertwined model of Gemini, Genie and something like Nano Banana. Nano Banana shows first climps of AGI "It also has Gemini under the hood, it sort of understands what's going on semantically in those images, so I think that's sort of, it's getting towards a kind of AGIfor imaging" AGI will be the unification of the Genie 3, Nano Banana and Gemini models "And then eventually, we got to kind of converge all of those different projects. And they're intertwined. But we need to, you know, converge them all into one" Google seems to be doing everything right at the moment. With Genie 3, Gemini 3, and Nano Banana, they have all the prerequisites to create a general model that develops a true understanding of the real world.

Chubby♨️

66,335 次观看 • 6 个月前