I was first introduced to neural networks as an undergraduate in 1990. Back then, many people in the AI community were excited about the potential of neural networks, which were impressive, but couldn’t yet accomplish important, real-world tasks. I was excited, too! I did my senior thesis on using parallel computation to train neural networks, thinking we only needed 32X more compute power to get there. I was way off. At that time, we needed 1 million times as much computational power.

A short 21 years later, with exponentially more computational power, it was time to take another crack at neural networks. In 2011, I and a few others at Google started training very large neural networks using millions of randomly selected frames from videos online. The results were remarkable. Without explicit training, the system automatically learned to recognize different objects (especially cats, the Internet is full of cats). This was one transformational discovery in AI among a long string of successes that is still ongoing — at Google and elsewhere.

I share my own history of neural networks to illustrate that, while progress in AI might feel especially fast right now, it’s come from a long arc of progress. In fact, prior to 2012, computers had a really difficult time seeing, hearing, or understanding spoken or written language. Over the past 10 years we’ve made especially rapid progress in AI.

Today, we’re excited about many recent advances in AI that Google is leading — not just on the technical side, but in responsibly deploying it in ways that help people around the world. That means deploying AI in Google Cloud, in our products from Pixel phones to Google Search, and in many fields of science and other human endeavors.

We’re aware of the challenges and risks that AI poses as an emerging technology. We were the first major company to release and operationalize a set of AI Principles, and following them has actually (and some might think counterintuitively) allowed us to focus on making rapid progress on technologies that can be helpful to everyone. Getting AI right needs to be a collective effort — involving not just researchers, but domain experts, developers, community members, businesses, governments and citizens.

I’m happy to make announcements in three transformative areas of AI today: first, using AI to make technology accessible in many more languages. Second, exploring how AI might bolster creativity. And third, in AI for Social Good, including climate adaptation.

1. Supporting 1,000 languages with AI

Language is fundamental to how people communicate and make sense of the world. So it’s no surprise it’s also the most natural way people engage with technology. But more than 7,000 languages are spoken around the world, and only a few are well represented online today. That means traditional approaches to training language models on text from the web fail to capture the diversity of how we communicate globally. This has historically been an obstacle in the pursuit of our mission to make the world’s information universally accessible and useful.

That’s why today we’re announcing the 1,000 Languages Initiative, an ambitious commitment to build an AI model that will support the 1,000 most spoken languages, bringing greater inclusion to billions of people in marginalized communities all around the world. This will be a many years undertaking – some may even call it a moonshot – but we are already making meaningful strides here and see the path clearly. Technology has been changing at a rapid clip – from the way people use it to what it’s capable of. Increasingly, we see people finding and sharing information via new modalities like images, videos, and speech. And our most advanced language models are multimodal – meaning they’re capable of unlocking information across these many different formats. With these seismic shifts come new opportunities.

As part of our this initiative and our focus on multimodality, we’ve developed a Universal Speech Model — or USM — that’s trained on over 400 languages, making it the largest language coverage seen in a speech model to date. As we expand on this work, we’re partnering with communities across the world to source representative speech data. We recently announced voice typing for 9 more African languages on Gboard by working closely with researchers and organizations in Africa to create and publish data. And in South Asia, we are actively working with local governments, NGOs, and academic institutions to eventually collect representative audio samples from across all the regions’ dialects and languages.

2. Empowering creators and artists with AI

AI-powered generative models have the potential to unlock creativity, helping people across cultures express themselves using video, imagery, and design in ways that they previously could not.

Our researchers have been hard at work developing models that lead the field in terms of quality, generating images that human raters prefer over other models. We recently shared important breakthroughs, applying our diffusion model to video sequences and generating long coherent videos for a sequence of text prompts. We can combine these techniques to produce video — for the first time, today we’re sharing AI-generated super-resolution video:

We’ll soon be bringing our text-to-image generation technologies to AI Test Kitchen, which provides a way for people to learn about, experience, and give feedback on emerging AI technology. We look forward to hearing feedback from users on these demos in AI Test Kitchen Season 2. You’ll be able to build themed cities with “City Dreamer” and design friendly monster characters that can move, dance, and jump with “Wobble” — all by using text prompts.

In addition to 2D images, text-to-3D is now a reality with DreamFusion, which produces a three-dimensional model that can be viewed from any angle and can be composited into any 3D environment. Researchers are also making significant progress in the audio generation space with AudioLM, a model that learns to generate realistic speech and piano music by listening to audio only. In the same way a language model might predict the words and sentences that follow a text prompt, AudioLM can predict which sounds should follow after a few seconds of an audio prompt.

We're collaborating with creative communities globally as we develop these tools. For example, we're working with writers using Wordcraft, which is built on our state-of-the-art dialog system LaMDA, to experiment with AI-powered text generation. You can read the first volume of these stories at the Wordcraft Writers Workshop.

3. Addressing climate change and health challenges with AI

AI also has great potential to address the effects of climate change, including helping people adapt to new challenges. One of the worst is wildfires, which affect hundreds of thousands of people today, and are increasing in frequency and scale.

Today, I’m excited to share that we’ve advanced our use of satellite imagery to train AI models to identify and track wildfires in real time, helping predict how they will evolve and spread. We’ve launched this wildfire tracking system in the U.S., Canada, Mexico, and are rolling out in parts of Australia, and since July we’ve covered more than 30 big wildfire events in the U.S. and Canada, helping inform our users and firefighting teams with over 7 million views in Google Search and Maps.

We’re also using AI to forecast floods, another extreme weather pattern exacerbated by climate change. We’ve already helped communities to predict when floods will hit and how deep the waters will get — in 2021, we sent 115 million flood alert notifications to 23 million people over Google Search and Maps, helping save countless lives. Today, we’re sharing that we’re now expanding our coverage to more countries in South America (Brazil and Colombia), Sub-Saharan Africa (Burkina Faso, Cameroon, Chad, Democratic Republic of Congo, Ivory Coast, Ghana, Guinea, Malawi, Nigeria, Sierra Leone, Angola, South Sudan, Namibia, Liberia, and South Africa), and South Asia (Sri Lanka). We’ve used an AI technique called transfer learning to make it work in areas where there’s less data available. We’re also announcing the global launch of Google FloodHub, a new platform that displays when and where floods may occur. We’ll also be bringing this information to Google Search and Maps in the future to help more people to reach safety in flooding situations.

Finally, AI is helping provide ever more access to healthcare in under-resourced regions. For example, we’re researching ways AI can help read and analyze outputs from low-cost ultrasound devices, giving parents the information they need to identify issues earlier in a pregnancy. We also plan to continue to partner with caregivers and public health agencies to expand access to diabetic retinopathy screening through our Automated Retinal Disease Assessment tool (ARDA). Through ARDA, we’ve successfully screened more than 150,000 patients in countries like India, Thailand, Germany, the United States, and the United Kingdom across deployed use and prospective studies — more than half of those in 2022 alone. Further, we’re exploring how AI can help your phone detect respiratory and heart rates. This work is part of Google Health’s broader vision, which includes making healthcare more accessible for anyone with a smartphone.

AI in the years ahead

Our advancements in neural network architectures, machine learning algorithms and new approaches to hardware for machine learning have helped AI solve important, real-world problems for billions of people. Much more is to come. What we’re sharing today is a hopeful vision for the future — AI is letting us reimagine how technology can be helpful. We hope you’ll join us as we explore these new capabilities and use this technology to improve people’s lives around the world.

When I reflect on the past two decades of computer science research, few things inspire me more than the remarkable progress we’ve seen in the field of artificial intelligence.

In 2001, some colleagues sitting just a few feet away from me at Google realized they could use an obscure technique called machine learning to help correct misspelled Search queries. (I remember I was amazed to see it work on everything from “ayambic pitnamiter” to “unnblevaiabel”). Today, AI augments many of the things that we do, whether that’s helping you capture a nice selfie, or providing more useful search results, or warning hundreds of millions of people when and where flooding will occur. Twenty years of advances in research have helped elevate AI from a promising idea to an indispensable aid in billions of people’s daily lives. And for all that progress, I’m still excited about its as-yet-untapped potential – AI is poised to help humanity confront some of the toughest challenges we’ve ever faced, from persistent problems like illness and inequality to emerging threats like climate change.

But matching the depth and complexity of those urgent challenges will require new, more capable AI systems – systems that can combine AI’s proven approaches with nascent research directions to be able to solve problems we are unable to solve today. To that end, teams across Google Research are working on elements of a next-generation AI architecture we think will help realize such systems.

We call this new AI architecture Pathways.

Pathways is a new way of thinking about AI that addresses many of the weaknesses of existing systems and synthesizes their strengths. To show you what I mean, let’s walk through some of AI’s current shortcomings and how Pathways can improve upon them.

Today's AI models are typically trained to do only one thing. Pathways will enable us to train a single model to do thousands or millions of things.

Today’s AI systems are often trained from scratch for each new problem – the mathematical model’s parameters are initiated literally with random numbers. Imagine if, every time you learned a new skill (jumping rope, for example), you forgot everything you’d learned – how to balance, how to leap, how to coordinate the movement of your hands – and started learning each new skill from nothing.

That’s more or less how we train most machine learning models today. Rather than extending existing models to learn new tasks, we train each new model from nothing to do one thing and one thing only (or we sometimes specialize a general model to a specific task). The result is that we end up developing thousands of models for thousands of individual tasks. Not only does learning each new task take longer this way, but it also requires much more data to learn each new task, since we’re trying to learn everything about the world and the specifics of that task from nothing (completely unlike how people approach new tasks).

Instead, we’d like to train one model that can not only handle many separate tasks, but also draw upon and combine its existing skills to learn new tasks faster and more effectively. That way what a model learns by training on one task – say, learning how aerial images can predict the elevation of a landscape – could help it learn another task -- say, predicting how flood waters will flow through that terrain.

We want a model to have different capabilities that can be called upon as needed, and stitched together to perform new, more complex tasks – a bit closer to the way the mammalian brain generalizes across tasks.

Today's models mostly focus on one sense. Pathways will enable multiple senses.

People rely on multiple senses to perceive the world. That’s very different from how contemporary AI systems digest information. Most of today’s models process just one modality of information at a time. They can take in text, or images or speech — but typically not all three at once.

Pathways could enable multimodal models that encompass vision, auditory, and language understanding simultaneously. So whether the model is processing the word “leopard,” the sound of someone saying “leopard,” or a video of a leopard running, the same response is activated internally: the concept of a leopard. The result is a model that’s more insightful and less prone to mistakes and biases.

And of course an AI model needn’t be restricted to these familiar senses; Pathways could handle more abstract forms of data, helping find useful patterns that have eluded human scientists in complex systems such as climate dynamics.

Today's models are dense and inefficient. Pathways will make them sparse and efficient.

A third problem is that most of today’s models are “dense,” which means the whole neural network activates to accomplish a task, regardless of whether it’s very simple or really complicated.

This, too, is very unlike the way people approach problems. We have many different parts of our brain that are specialized for different tasks, yet we only call upon the relevant pieces for a given situation. There are close to a hundred billion neurons in your brain, but you rely on a small fraction of them to interpret this sentence.

AI can work the same way. We can build a single model that is “sparsely” activated, which means only small pathways through the network are called into action as needed. In fact, the model dynamically learns which parts of the network are good at which tasks -- it learns how to route tasks through the most relevant parts of the model. A big benefit to this kind of architecture is that it not only has a larger capacity to learn a variety of tasks, but it’s also faster and much more energy efficient, because we don’t activate the entire network for every task.

For example, GShard and Switch Transformer are two of the largest machine learning models we’ve ever created, but because both use sparse activation, they consume less than 1/10th the energy that you’d expect of similarly sized dense models — while being as accurate as dense models.

So to recap: today’s machine learning models tend to overspecialize at individual tasks when they could excel at many. They rely on one form of input when they could synthesize several. And too often they resort to brute force when deftness and specialization of expertise would do.

That’s why we’re building Pathways. Pathways will enable a single AI system to generalize across thousands or millions of tasks, to understand different types of data, and to do so with remarkable efficiency – advancing us from the era of single-purpose models that merely recognize patterns to one in which more general-purpose intelligent systems reflect a deeper understanding of our world and can adapt to new needs.

That last point is crucial. We’re familiar with many of today’s biggest global challenges, and working on technologies to help address them. But we’re also sure there are major future challenges we haven’t yet anticipated, and many will demand urgent solutions. So, with great care, and always in line with our AI Principles, we’re crafting the kind of next-generation AI system that can quickly adapt to new needs and solve new problems all around the world as they arise, helping humanity make the most of the future ahead of us.