Google AI: Release Notes

Chapters:
00:00 - Intro and defining world models and RL roots
01:51 - Demo: Goldfish and shark in underwater world
04:59 - Project Genie gallery
06:31 - Physics, remixing, and UI prompts
11:00 - Demo: Nano Banana mascot “Bob”
13:20 - Constraints, generation limits, and infrastructure
17:04 - Trusted testers and robotics future
28:34 - Frontier prompting and universal simulation
29:27 - Cross-Google collaboration
31:16 - Adoption timelines and impact
34:16 - Model generalization and historical context
38:52 - Hardware limits and the slope of progress

Rhiannon Bell and Robby Stein, Product and Design leads for Google Search, join host Logan Kilpatrick for a deep dive into the integration of Gemini 3 into Search. Their conversation explores the evolution of Generative UI, where models act as designers to create bespoke, interactive simulations on the fly. Learn more about the role of Gemini 3 Flash in delivering speed at scale, the development of Search's new "persona," and how models like Nano Banana are powering next-generation data visualization.

Watch on YouTube: https://www.youtube.com/watch?v=AqyclkRBSe4

Chapters:
0:00 - Introduction
1:24 - What is Generative UI?
2:23 - From static to generative design
6:37 - Interactive simulations
8:47 - Latency and visual QA
10:48 - Gemini 3 Flash in Search
12:08 - Fusing AI Mode and AI Overviews
14:24 - The Search persona
17:12 - Agentic system understanding
18:22 - Visualizing data with Nano Banana

Logan Kilpatrick from Google DeepMind sits down with Sundar Pichai, CEO of Google and Alphabet to discuss the launch of Gemini 3, Nano Banana Pro and Google's overall AI momentum. They talk about Google’s long-term bets on infrastructure, what it’s actually like to ship SOTA models, and the rise of vibe coding. Sundar also shares his personal launch day rituals and thoughts on future moonshots like putting data centers in space.

Watch on YouTube: https://www.youtube.com/watch?v=iFqDyWFuw1c

Chapters:
0:00 - Intro
0:51 - Shipping Gemini 3
2:44 - Google's decade-long investment in AI
4:27 - The full stack advantage
5:43 - Scaling up compute and capacity
7:32 - Sim-shipping Gemini across products
9:35 - Nano Banana Pro
12:13 - Monitoring launch day
14:13 - Future model roadmap
16:05 - Launch day rituals
18:02 - The Blue Micro Kitchen
21:57 - Future moonshots
23:26 - The rise of vibe coding
26:50 - What’s next

Introducing Nano Banana Pro, a powerful model built on Gemini 3 Pro, designed to enhance text rendering, infographics, and structured content generation. Tune in to learn about Nano Banana Pro’s advanced visual reasoning and multi-turn generation capabilities, and how this next-gen tool enables complex image edits and real-world applications. In this episode, we discuss how user feedback and continuous benchmarking drive model improvements, ensuring a superior experience for developers.

Watch on YouTube: https://www.youtube.com/watch?v=hk6gwiZmSWA

Chapters:
00:00 - Introducing Nano Banana Pro
02:00 - Enhanced world understanding
04:59 - Advanced text rendering
05:49 - Gemini 3 Pro's influence
09:30 - Multi-turn & infographics
14:04 - Text rendering comparison
16:26 - Multilingual text support
18:22 - Infographics for learning
24:00 - Multi-image input
26:38 - Resolution & fidelity
30:07 - Advanced editing & style
32:09 - Practical use cases
35:26 - Future outlook & thanks

Join Logan Kilpatrick and Koray Kavukcuoglu, CTO of Google DeepMind and Chief AI Architect of Google, as they discuss Gemini 3 and the state of AI!

Their conversation includes the reception of Gemini 3, the ongoing advancements in AI research, and the role of benchmarks in pushing new frontiers. They explore critical areas for Gemini's focus, emphasizing instruction following, tool calls, and internationalization, alongside Google's collaborative approach to AI development.

Watch on YouTube: https://www.youtube.com/watch?v=fXtna7UrL44

Chapters:
0:00 - Intro
2:00 - Gemini 3 launch reception
4:16 - Continuous progress and innovation
6:47 - Key areas for Gemini improvement
11:45 - Product scaffolding for model improvement
13:56 - Chief AI architect role
17:04 - Engineering mindset and collaboration
18:37 - Future growth areas for Gemini
20:33 - From research to engineering mindset
23:22 - The rise of generative media
27:22 - Nano Banana Pro capabilities
29:31 - Towards unified model checkpoints
36:26 - Organizing for AI success
38:26 - Balancing exploration and scaling
41:40 - DeepMind's collaborative culture
45:21 - Innovating at Google
48:37 - Closing

Explore Antigravity, Google DeepMind’s innovative new AI developer coding product, with Varun Mohan on Release Notes. This episode dives into Antigravity as a powerful agent development platform, integrating a familiar IDE experience with browser verification and Gemini 3.0 capabilities. Discover how developers can orchestrate complex agentic workflows, leverage artifacts for task communication, and balance AI automation with human collaboration. Learn about the philosophy behind building next-gen agentic experiences, the platform's multimodal strengths, and its role in accelerating software development at scale.

Watch on YouTube: https://www.youtube.com/watch?v=uzFOhkORVfk

Chapters
00:00 - Introducing Google Antigravity
04:02 - Evolution of AI in coding
04:53 - Beyond writing code
06:21 - Ideal Google Antigravity user
09:48 - Evolving user personas
11:46 - Agents versus the IDE
14:46 - Human-agent collaboration
16:43 - Local versus server-side
18:50 - Self-improvement and knowledge
21:29 - Generalizing agent capabilities
24:20 - Naming Google Antigravity
27:04 - Integrating Google's AI models
27:59 - Demo: Airbnb for dogs
28:48 - Understanding artifacts
29:51 - Asynchronous user feedback
32:16 - Agent manager workflow
33:17 - Browser actuation demo
34:36 - Browser for research and testing
36:45 - Parallel agent conversations
41:04 - Agent task best practices
42:51 - Future of Google Antigravity

Join us for a special episode of Release Notes as we unpack Gemini 3, Google’s latest AI model with key team members. Learn how Gemini 3 empowers developers with enhanced multimodal understanding, agentic capabilities for complex tasks, and generative interfaces that transform prompts into interactive applications. We discuss real-world use cases, the iterative development process driven by user feedback, and the strategic balance between model performance and broad accessibility across various Google platforms.

Watch on YouTube: https://www.youtube.com/watch?v=mci0f2dy7G0

Chapters:
00:00 - Introducing Gemini 3
03:08 - Gemini 3 everywhere
04:13 - The product-model partnership
08:20 - Balancing speed and quality
11:40 - Gemini 3 'wow' moments
27:47 - Generative interfaces and UI
31:44 - Gemini's agentic capabilities
33:55 - Proactive AI and future
34:55 - Managing compute demand
39:32 - The Gemini 3 family
41:45 - Conclusion

Dumi Erhan, co-lead of the Veo project at Google DeepMind, joins host Logan Kilpatrick for a deep dive into the evolution of generative video models. They discuss the journey from early research in 2018 to the launch of state-of-the-art Veo 3 model with native audio generation. Learn about the technical hurdles in evaluating and scaling video models, the challenges of long-duration video coherence and how user feedback is shaping the future of AI-powered video creation.

Chapter: 
0:00 - Intro
0:47 - Veo project's beginnings
3:02 - Veo's origins in Google Brain
5:07 - Video prediction and robotics applications
7:45 - Early progress and evaluation challenges
10:30 - Physics-based evaluations and their limitations
12:18 - The launch of the original Veo model
14:06 - Scaling challenges for video models
16:02 - The leap from Veo1 to Veo2
19:40 - Veo 3’s viral audio moment
21:17 - User trends shaping Veo's roadmap
23:49 - Image-to-video vs. text-to-video complexity
26:00 - New prompting methods and user control
27:55 - Coherence in long video generation
31:03 - Genie 3 and world models
35:54 - The steerability challenge
41:59 - Capability transfer and image data's role
47:25 - Closing

Pushmeet Kohli, Head of Science and Strategic Initiatives at Google DeepMind, joins host Logan Kilpatrick to explore the intersection of AI and scientific discovery. Learn how the team's unique problem-solving framework led to innovations like AlphaFold and AlphaEvolve, and how new tools like AI Co-scientist aim to democratize these types of breakthroughs for everyone. 

Watch on YouTube: https://www.youtube.com/watch?v=o7mdsL6BHsk

Chapters: 
0:00 - Intro
1:04 - Recent Alpha launches
02:15 - Framework for selecting research domains
06:21  - Scientific, commercial and social impact
15:00 - Wielding AGI for breakthroughs
16:48 - Tech transfer and team collaboration
19:46  -  IMO Gold Medal
21:42  - Evaluating math proofs
22:55 - From specialized models to Deep Think
24:22 -  Do math skills generalize?
25:53 - Generalizing the IMO model
27:43 - Democratizing AI science tools
30:09 - AI Co-scientist
35:17 - An API for science?

Join host Logan Kilpatrick in discussion with some of the minds behind Google's new state-of-the-art image model, Gemini 2.5 Flash. Product and research leads from the Gemini team break down the technology behind its key capabilities, including interleaved generation for complex edits and new approaches to achieving character consistency and pixel-perfect control. With Nicole Brichtova, Kaushik Shivakumar, Mostafa Dehghani and Robert Riachi. 

Watch on YouTube: 

Chapters:
0:37 - New model introduction
1:21 -Demo - Image Editing
3:44 - Text rendering capabilities
4:44 Beyond human preference evals
6:44 - Text rendering as a proxy for quality
8:38 - Positive transfer between modalities
11:25 - Demo - Multi-turn, context aware image generation
13:54 - Pixel-perfect editing and character consistency
15:51 - Interleaved image generation
17:59 - Specialized vs. native models
19:52 - Understanding nuanced prompts
20:59 - User feedback shaping model development
22:37 - Improvements in character consistency
24:17 - More natural looking images from team collaboration
26:41 - What’s next for image generation models