Empowering YouTube creators with generative AI - Related to healthcare, youtube, tools, reliable, our

Developing reliable AI tools for healthcare

New research proposes a system to determine the relative accuracy of predictive AI in a hypothetical medical setting, and when the system should defer to a human clinician.

Artificial intelligence (AI) has great potential to enhance how people work across a range of industries. But to integrate AI tools into the workplace in a safe and responsible way, we need to develop more robust methods for understanding when they can be most useful.

So when is AI more accurate, and when is a human? This question is particularly essential in healthcare, where predictive AI is increasingly used in high-stakes tasks to assist clinicians.

Today in Nature Medicine, we’ve , which proposes CoDoC (Complementarity-driven Deferral-to-Clinical Workflow), an AI system that learns when to rely on predictive AI tools or defer to a clinician for the most accurate interpretation of medical images.

CoDoC explores how we could harness human-AI collaboration in hypothetical medical settings to deliver the best results. In one example scenario, CoDoC reduced the number of false positives by 25% for a large, de-identified UK mammography dataset, compared with commonly used clinical workflows – without missing any true positives.

This work is a collaboration with several healthcare organisations, including the United Nations Office for Project Services’ Stop TB Partnership. To help researchers build on our work to improve the transparency and safety of AI models for the real world, we’ve also open-sourced CoDoC’s code on GitHub.

CoDoC: Add-on tool for human-AI collaboration.

Building more reliable AI models often requires re-engineering the complex inner workings of predictive AI models. However, for many healthcare providers, it’s simply not possible to redesign a predictive AI model. CoDoC can potentially help improve predictive AI tools for its consumers without requiring them to modify the underlying AI tool itself.

When developing CoDoC, we had three criteria:

Non-machine learning experts, like healthcare providers, should be able to deploy the system and run it on a single computer.

Training would require a relatively small amount of data – typically, just a few hundred examples.

The system could be compatible with any proprietary AI models and would not need access to the model’s inner workings or data it was trained on.

Determining when predictive AI or a clinician is more accurate.

With CoDoC, we propose a simple and usable AI system to improve reliability by helping predictive AI systems to ‘know when they don’t know’. We looked at scenarios, where a clinician might have access to an AI tool designed to help interpret an image, for example, examining a chest x-ray for whether a tuberculosis test is needed.

For any theoretical clinical setting, CoDoC’s system requires only three inputs for each case in the training dataset.

The predictive AI outputs a confidence score between 0 (certain no disease is present) and 1 (certain that disease is present). The clinician’s interpretation of the medical image. The ground truth of whether disease was present, as, for example, established via biopsy or other clinical follow-up.

Note: CoDoC requires no access to any medical images.

Research RT-2: New model translates vision and language into action Share.

Robotic Transformer 2 (RT-2) is a novel vision-language-...

Real-time communication is everywhere – live chatbots, data streams, or instant messaging. WebSockets are a powerful ena...

Research Mastering Stratego, the classic game of imperfect information Share.

DeepNash learns to play Stratego from scratch by comb...

Empowering YouTube creators with generative AI

Throughout this journey, we’ve worked closely with artists and creators and have been guided by their curiosity and feedback to ensure our technologies help more people realize their creative vision.

From our groundbreaking Transformer architecture to years of research in diffusion models, these models are built on nearly a decade of innovation at Google and are now optimized for wide-scale use.

We’re changing that, and making these incredible technologies more easily accessible to millions of creators and billions of customers around the world. Over the next few months, we’re bringing our advanced generative AI models, Veo and Imagen 3 , to YouTube creators through Dream Screen .

Artificial intelligence (AI) technologies for generating creative content are improving rapidly, but seamless ways of using them still aren’t widely available.

New video generation technology in YouTube Shorts will help millions of people realize their creative vision.

In 2025, creators will also be able to use Dream Screen for generating 6 second standalone video clips for their Shorts, like a cinematic underwater reveal of the Golden Gate Bridge.

Starting with an initial text prompt, Dream Screen uses Imagen 3 to generate four different images. Creators can pick an image in their preferred style, composition or aesthetic from these options. With the selected image, Veo generates a high-quality 6 second background video, tailored to the user’s creative vision.

By integrating Veo into Dream Screen, YouTube creators will soon be able to generate exciting backgrounds for their Shorts. Early next year, we’re also making it possible to generate 6 second standalone video clips.

We hope these new video and music generation technologies will inspire more people to bring their ideas to life in vivid, transformative ways.

To help identify AI-generated content, these creations will be watermarked using SynthID, and YouTube will apply a label that clearly communicates to viewers that it was generated with AI.

We’re continuing to improve our generative AI models and look forward to helping build technologies that unlock creativity for everyone, everywhere.

Earlier this year, we introduced our video generation model, Veo, and our latest image generation model, Imagen 3. Since then, it’s been exciting to w...

A note from Google and Alphabet CEO Sundar Pichai:

Last week, we rolled out our most capable model, Gemini [website] Ultra, and took a significant step for...

Research Building interactive agents in video game worlds Share.

Introducing a framework to create AI agents that can understand hu...

The next chapter of our Gemini era

In December, we took a significant step on our journey to make AI more helpful for everyone with the start of the Gemini era, setting a new state of the art across a wide range of text, image, audio, and video benchmarks. However, Gemini is evolving to be more than just the models. It supports an entire ecosystem — from the products that billions of people use every day, to the APIs and platforms helping developers and businesses innovate.

The largest model Ultra [website] is the first to outperform human experts on MMLU (massive multitask language understanding), which uses a combination of 57 subjects — including math, physics, history, law, medicine and ethics — to test knowledge and problem-solving abilities.

Today we’re taking our next step and bringing Ultra to our products and the world.

Bard has been the best way for people to directly experience our most capable models. To reflect the advanced tech at its core, Bard will now simply be called Gemini. It’s available in 40 languages on the web, and is coming to a new Gemini app on Android and on the Google app on iOS.

The version with Ultra will be called Gemini Advanced, a new experience far more capable at reasoning, following instructions, coding, and creative collaboration. For example, it can be a personal tutor, tailored to your learning style. Or it can be a creative partner, helping you plan a content strategy or build a business plan. You can read more in this post.

You can start using Gemini Advanced by subscribing to the new Google One AI Premium plan, which offers the best of Google’s AI elements in a single place. This premium plan builds off the popular Google One service offering expanded storage and .

Impact Stopping malaria in its tracks Share.

Developing a improved malaria vaccine with the help of AI that could save hundreds of th...

corporation 2023: A Year of Groundbreaking Advances in AI and Computing Share.

This has been a year of incredible progress in the field...

Technologies Watermarking AI-generated text and video with SynthID Share.

Announcing our novel watermarking method for AI-generated...