Uploaded December 2024 | Updated November 2026, 2 days ago
Google DeepMind has unveiled significant updates to its Veo 2 video generation model and Imagen 3 image generation model, alongside a new experiment called Whisk via its Google Labs Experiments.
Announcement: https://blog.google/technology/google-labs/video-image-generation-update-december-2024/
Google Labs Experiments: https://labs.google/experiments
00:00 Google DeepMind Video & Image Model Updates
01:20 Accessing Veo 2 & Imagen 3 Tools
02:07 ImageFX & Imagen 3 Tests
05:11 Testing Whisk Tool to Combine Photo-Realistic Images
10:04 Combining 2 Visual Styles Into One Image
These advancements represent a major leap forward in AI-powered media creation, offering users unprecedented control and quality in generating visual content. Veo 2 builds upon its predecessor, delivering enhanced capabilities in video generation, while Imagen 3 pushes the boundaries of image creation with improved detail and versatility.
Veo 2, Google's state-of-the-art video generation model, now produces high-quality videos up to 4K resolution and can extend to minutes in length[1][6]. The model demonstrates an improved understanding of real-world physics and human movement, resulting in more realistic and detailed outputs[1]. Veo 2 excels in following complex instructions, allowing users to specify genres, lenses, and cinematic effects with remarkable accuracy[1][6]. This level of control enables creators to craft videos with specific camera angles, lighting, and other cinematographic elements, opening up new possibilities for content creation across various industries.
Imagen 3, Google's latest image generation model, offers significant improvements in output quality and versatility[1][7]. The model generates brighter, better-composed images with richer details and textures, capable of rendering diverse art styles with greater accuracy[1]. Imagen 3's enhanced prompt understanding allows it to capture nuances like specific camera angles or compositions in long, complex prompts, making it easier for users to achieve desired results without extensive prompt engineering[7]. The model's improved text rendering capabilities also expand its potential use cases, from creating stylized birthday cards to designing presentations[7].
Alongside these updates, Google Labs has introduced Whisk, a new experiment that leverages the updated Imagen 3 and Gemini's visual understanding capabilities. Whisk allows users to remix existing images using prompts, showcasing the advancements in AI-powered image manipulation. These tools are being rolled out gradually, with Veo 2 capabilities coming to VideoFX and plans to expand to YouTube Shorts and other products next year[1]. Imagen 3 is now globally available in ImageFX, Google's image generation tool, in more than 100 countries[1].
Citations:
[1] https://blog.google/technology/google-labs/video-image-generation-update-december-2024/
[2] https://deepmind.google/technologies/veo/veo-2/
[3] yahoo.com/tech/google-deepmind-veo-2-ai-170100616.html
[4] cloud.google.com/vertex-ai/generative-ai/docs/image/model-versioning
[5]
[6] infoq.com/news/2024/12/veo-imagen-vertex-ai
[7] 9to5google.com/2024/12/16/google-veo-2
[8] https://deepmind.google/technologies/imagen-3/
[9] mashable.com/article/google-deepmind-veo-2-imagen-3-launch
[10] https://blog.google/products/gemini/google-gemini-update-august-2024/
[11] techcrunch.com/2024/12/16/google-deepmind-unveils-a-new-video-model-to-rival-sora
[12] cloud.google.com/blog/products/ai-machine-learning/introducing-veo-and-imagen-3-on-vertex-ai
[13] androidcentral.com/apps-software/google-ai-whisk-image-video-generation-updates
----------
š More templates and resources for tracking and reporting, including a Campaign URL Builder: build.growthlearner.com
Learn more insights on measuring website traffic, analysis, and reporting, including GA4 and other tools: growthlearner.com/?utm_source=youtube&utm_medium=social
Google DeepMind has unveiled significant updates to its Veo 2 video generation model and Imagen 3 image generation model, alongside a new experiment called Whisk via its Google Labs Experiments.
Announcement: https://blog.google/technology/google-labs/video-image-generation-update-december-2024/
Google Labs Experiments: https://labs.google/experiments
00:00 Google DeepMind Video & Image Model Updates
01:20 Accessing Veo 2 & Imagen 3 Tools
02:07 ImageFX & Imagen 3 Tests
05:11 Testing Whisk Tool to Combine Photo-Realistic Images
10:04 Combining 2 Visual Styles Into One Image
These advancements represent a major leap forward in AI-powered media creation, offering users unprecedented control and quality in generating visual content. Veo 2 builds upon its predecessor, delivering enhanced capabilities in video generation, while Imagen 3 pushes the boundaries of image creation with improved detail and versatility.
Veo 2, Google's state-of-the-art video generation model, now produces high-quality videos up to 4K resolution and can extend to minutes in length[1][6]. The model demonstrates an improved understanding of real-world physics and human movement, resulting in more realistic and detailed outputs[1]. Veo 2 excels in following complex instructions, allowing users to specify genres, lenses, and cinematic effects with remarkable accuracy[1][6]. This level of control enables creators to craft videos with specific camera angles, lighting, and other cinematographic elements, opening up new possibilities for content creation across various industries.
Imagen 3, Google's latest image generation model, offers significant improvements in output quality and versatility[1][7]. The model generates brighter, better-composed images with richer details and textures, capable of rendering diverse art styles with greater accuracy[1]. Imagen 3's enhanced prompt understanding allows it to capture nuances like specific camera angles or compositions in long, complex prompts, making it easier for users to achieve desired results without extensive prompt engineering[7]. The model's improved text rendering capabilities also expand its potential use cases, from creating stylized birthday cards to designing presentations[7].
Alongside these updates, Google Labs has introduced Whisk, a new experiment that leverages the updated Imagen 3 and Gemini's visual understanding capabilities. Whisk allows users to remix existing images using prompts, showcasing the advancements in AI-powered image manipulation. These tools are being rolled out gradually, with Veo 2 capabilities coming to VideoFX and plans to expand to YouTube Shorts and other products next year[1]. Imagen 3 is now globally available in ImageFX, Google's image generation tool, in more than 100 countries[1].
Citations:
[1] https://blog.google/technology/google-labs/video-image-generation-update-december-2024/
[2] https://deepmind.google/technologies/veo/veo-2/
[3] yahoo.com/tech/google-deepmind-veo-2-ai-170100616.html
[4] cloud.google.com/vertex-ai/generative-ai/docs/image/model-versioning
[5]
[6] infoq.com/news/2024/12/veo-imagen-vertex-ai
[7] 9to5google.com/2024/12/16/google-veo-2
[8] https://deepmind.google/technologies/imagen-3/
[9] mashable.com/article/google-deepmind-veo-2-imagen-3-launch
[10] https://blog.google/products/gemini/google-gemini-update-august-2024/
[11] techcrunch.com/2024/12/16/google-deepmind-unveils-a-new-video-model-to-rival-sora
[12] cloud.google.com/blog/products/ai-machine-learning/introducing-veo-and-imagen-3-on-vertex-ai
[13] androidcentral.com/apps-software/google-ai-whisk-image-video-generation-updates
----------
š More templates and resources for tracking and reporting, including a Campaign URL Builder: build.growthlearner.com
Learn more insights on measuring website traffic, analysis, and reporting, including GA4 and other tools: growthlearner.com/?utm_source=youtube&utm_medium=social


![ChatGPT Canvas Update: Design & Create Apps with Code Rendering
ChatGPT Canvas released an update to enable developers to create and design apps and web pages, and render code. This powerful feature allows users to collaborate with AI on complex coding projects, including HTML and React development, directly within the ChatGPT interface.
00:00 Rendering Code/App in ChatGPT Canvas
01:49 Optimizing Your App & Coding with ChatGPT Canvas
03:18 Iterating & Interacting/Testing with Your App
05:17 Creating Web Pages with ChatGPT Canvas
07:04 Comparison to AI Coding Assistants
Canvas provides a dedicated workspace for coding, app design, and web page design, offering a more intuitive and efficient environment compared to the traditional chat interface. Developers can now write, edit, and preview code in real-time, streamlining the development process.
The ability to render HTML and React code within Canvas enables instant visualization of changes, reducing the need for constant switching between development environments. This feature is particularly beneficial for front-end developers working on user interfaces and interactive components.
For app creation and design, Canvas serves as a versatile tool that supports various stages of the development lifecycle. From initial concept sketches to detailed prototypes, developers can leverage ChatGPTs AI capabilities to generate ideas, refine designs, and troubleshoot issues.
The side-by-side layout of Canvas facilitates seamless collaboration between the developer and AI, allowing for quick iterations and improvements. Additionally, Canvas supports a wide range of programming languages and frameworks, making it suitable for diverse app development projects across different platforms.
By integrating AI-assisted coding with real-time rendering, ChatGPT Canvas empowers developers to create more sophisticated and polished applications in less time, potentially accelerating time-to-market for new software products[1][3][5].
Citations:
- https://lwaltd.com/blog/how-businesses-can-benefit-from-chatgpts-canvas-feature
- https://openai.com/index/introducing-canvas/
- https://www.zdnet.com/article/im-a-chatgpt-power-user-heres-why-canvas-is-its-best-productivity-feature/
- https://zapier.com/blog/chatgpt-canvas/
- https://freshvanroot.com/blog/chatgpt-canvas-writers/
- https://learnprompting.org/blog/how-to-use-openai-canvas-chatgpt
Learn more insights on measuring website traffic, analysis, and reporting, including GA4 and other tools: https://growthlearner.com/?utm_source=youtube&utm_medium=social ChatGPT Canvas Update: Design & Create Apps with Code Rendering](https://i.ytimg.com/vi/Z7gUiUyZslA/mqdefault.jpg)

![Recapping the Biggest AI Updates
Googles Gemini 2.0 Flash experimental model, available through AI Studio, introduces groundbreaking stream realtime and multimodal capabilities. This experimental AI model enhances user interaction with software and websites through its real-time screen-sharing feature. Gemini 2.0 Flash represents a significant advancement in AI-assisted computing, offering users a more intuitive and responsive experience when navigating digital interfaces.
00:00 Intro
00:40 Gemini 2.0 Flash Explains AI Studio via Screen Share
02:44 Gemini 2.0 Flash Explains Code & Colab Notebook
03:51 Gemini 2.0 Flash Teaches Canva for Graphic Design
08:29 Gemini 2.0 Flash Teaches Jira
The experimental models real-time streaming capabilities allow it to process and respond to live video and audio input, making it an ideal tool for assisting users with software applications and website navigation. Gemini 2.0 Flash can analyze screen content in real-time, providing contextual assistance and suggestions as users interact with various digital platforms. This feature is particularly beneficial for tasks such as learning new software, troubleshooting technical issues, or guiding users through complex web interfaces. The models ability to understand and respond to multimodal inputs, including text, images, and video, enables it to offer comprehensive support across a wide range of digital environments[2].
Gemini 2.0 Flashs screen-sharing feature opens up new possibilities for AI-assisted productivity and learning. For example, the model can act as a live coding tutor, providing real-time feedback and suggestions as users write code[4]. It can also assist in generating code from screen captures, automating web testing processes. The models advanced capabilities extend to helping visually impaired users navigate websites and applications by describing on-screen elements and providing audio guidance. As an experimental AI model, Gemini 2.0 Flash represents a significant step towards more interactive and responsive AI assistants, potentially revolutionizing how users interact with digital interfaces and learn new software skills[1][2].
Citations:
[1] https://blog.google/products/gemini/google-gemini-deep-research/
[2] https://blog.google/technology/google-deepmind/google-gemini-ai-update-december-2024/
[3] https://cloud.google.com/vertex-ai/generative-ai/docs/gemini-v2
[4] https://developers.googleblog.com/en/the-next-chapter-of-the-gemini-era-for-developers/
[5] https://deepmind.google/technologies/gemini/flash/
[6] https://venturebeat.com/ai/google-gemini-2-0-flash-test-drive-reveals-why-every-analyst-needs-to-know-this-modelgoogle-gemini-2-0-flash-test-drive-why-every-analyst-needs-to-know-this-model/
[7] https://www.deeplearning.ai/the-batch/google-introduces-gemini-2-0-flash-a-faster-more-capable-ai-model/
[8] https://venturebeat.com/ai/gemini-2-0-flash-ushers-in-a-new-era-of-real-time-multimodal-ai/
[9] https://ai.google.dev/gemini-api/docs/thinking-mode
[10] https://ai.google.dev/gemini-api/docs/models/gemini-v2
š More templates and resources for tracking and reporting, including a Campaign URL Builder: https://build.growthlearner.com/
Learn more insights on measuring website traffic, analysis, and reporting, including GA4 and other tools: https://growthlearner.com/?utm_source=youtube&utm_medium=social Recapping the Biggest AI Updates](https://i.ytimg.com/vi/Zfn99yEfR90/mqdefault.jpg)





