Google DeepMind's GenCast AI Model Revolutionizes Weather Forecasting Accuracy

How AI is Revolutionizing Weather Forecasting

The Future of Weather Predictions: AI's Role in Revolutionizing Forecasting Artificial Intelligence (AI) is now rapidly becoming a transformative force across industries, including weather forecasting. Google DeepMind's new GenCast AI model exemplifies this potential, harnessing decades of climate data to significantly enhance predictive accuracy for weather events. Recently, Google DeepMind's unveiled a brand-new GenCast AI model -- which holds the potential to revolutionize the standards of reliability and precision related to weather forecasts. Originally developed from 39-years of climate data (1979-2018), GenCast had the challenging role of forecasting the climate for the year 2019. These were then compared with its predictions against other conventional forecasting engines including the Ensemble Forecast and real weather. The results were nothing short of extraordinary. GenCast outperformed the Ensemble Forecast in 97.2% of cases. For forecasts extending beyond 36 hours, its accuracy climbed even higher to an impressive 99.8%. Such precision makes GenCast a powerful tool for both routine forecasts and more critical predictions involving extreme weather events like tornadoes and hurricanes. An area that holds a lot of potential for GenCast, and consequently its utility, is the ability to enhance the reaction to hazardous weather conditions. If used in 'real time', its potential would be in identifying adverse occurrences early enough and the ability of the authorities to give alerts and organize evacuations faster, possibly reducing considerable losses of lives. Nonetheless, although AI has stepped up against these challenges, they cannot rule out the services of meteorologists. Determining the quality of the AI models such as GenCast depends on quality data and this must be gathered, analyzed and validated by experts. What human intelligence companionship expects to see in lieu of AI is a supplement to the weather predictive platform. Google's advances in AI don't stop at weather forecasting. The company also offers tools like Gemini AI, designed for productivity and integration across platforms like Google Workspace and PowerPoint. These innovations showcase the versatility of AI in addressing diverse challenges and improving everyday life. In conclusion, the AI has moved to a new step in weather forecasting as it declared the result within a short span that has failure rates higher than the terrestrial units it replaced. Some models like GenCast state that they can accurately give the long range and severe range of the weather prediction. This means that to add the human input to it along with the artificial intelligence, the weather prediction will continue to grow.

Google's weather tool gets it right 97% more times than best forecaster

Weather forecasting is about to get a whole lot more accurate thanks to Google DeepMind's GenCast AI-powered prediction model GenCast, an AI-powered weather prediction tool, can crank out 15 days of highly accurate weather forecasts in minutes. Not only will it help you dress right in the morning, but it can give valuable, life-saving warnings for extreme weather events. How many times have you checked the weather forecast then left the house prepared for a sunny day only to be ambushed by an afternoon storm? Or worn your heavy coat to find that the temperature climbs too high to keep it on? The fact is, even though modern meteorology has come a long way, the sheer complexity of using hundreds of data points to predict the weather makes accurate forecasts hard to come by - especially the further out in time you go. A new system recently announced by Google DeepMind called GenCast is set to make things a lot better. The AI-powered program was trained on four decades of historical data through 2018, taken from the European Centre for Medium-Range Weather Forecasts' (ECMWF) historical archives. The ECMWF is considered the most accurate weather-prediction service in the world. Archival data from the service including wind speed, temperature, and pressure at different altitudes all went into the training program. Once the AI model was trained up, it was then asked to predict the weather for 15-day periods in 2019. When compared to the actual weather that occurred in that year, GenCast was 97.2% more accurate than the ECMWF predictions. When the window of the prediction time was narrowed to just 36 hours, GenCast did even better - it was 99.8% more accurate than ECMWF. Both metrics best Google's previous weather prediction program, known as GraphCast. The new system works by generating 50 or more predictions of what the weather might look like based on current weather trends and then aggregates the information to make its forecast. According to Google, the system can generate its 15-day forecast in just eight minutes using one of the company's TPU v5 AI processors, as opposed to the hours it takes supercomputers to do the same. "GenCast is a diffusion model, the type of generative AI model that underpins the recent, rapid advances in image, video and music generation," says Google on its DeepMind blog. "However, GenCast differs from these, in that it's adapted to the spherical geometry of the Earth, and learns to accurately generate the complex probability distribution of future weather scenarios when given the most recent state of the weather as input." In addition to helping you dress right for the day's weather, Google says GenCast will also be able to save lives by helping predict the path of severe weather events - an increasing occurrence due to accelerating climate warming - days before they strike. The company says the data GenCast produces could also help renewable energy efforts by, for example, detecting wind patterns for optimal placement and usage of wind farms. While you can't currently download an app using GenCast, Google is making the data gleaned from testing as well as real-time forecasts available publicly and is encouraging researchers, meteorologists and other entities to take advantage of the technology, so don't be surprised if your current weather app starts become more accurate in the coming months.

Google Deepmind's new AI weather forecaster blows away the competition

GenCast predicts the weather faster than you can find your umbrella Google Deepmind researchers have built an AI weather forecasting tool that makes faster and more accurate predictions than the best system available today. Dubbed GenCast, the new model outperformed the ENS forecast, widely regarded as the world leader, 97% of the time for predictions up to 15 days in advance. It was tested on over 1,320 weather scenarios, including tropical cyclones and heatwaves. "Outperforming ENS marks something of an inflection point in the advance of AI for weather prediction," Ilan Price, a research scientist at Google DeepMind, told the Guardian. "At least in the short term, these models are going to accompany and be alongside existing, traditional approaches." GenCast is a diffusion machine learning model, similar to those used in generative AI for tasks like image or text creation. However, it's uniquely adapted for weather prediction, trained on four decades of data from the European Centre for Medium-Range Weather Forecasts (ECMWF) -- the agency behind ENS. During the experiments, researchers asked GenCast to generate a forecast for 2019. They then compared the results to the actual weather during that year as well as ENS' predictions. GenCast creates an ensemble of 50+ different predictions, each showing a possible future scenario. This data helps authorities prepare for extreme weather events like hurricanes or wind farm operators better predict power output days in advance. The fancy name for this technique is probabilistic ensemble forecasting. It's already the gold standard in traditional forecast systems. However, GenCast is taking things up a notch. The system can spit out predictions in far less time: 8 minutes, compared to hours for traditional models. That's because models like ENS run on massive supercomputers that have to crunch through millions of equations to make a prediction. By contrast, GenCast runs on a single Google Cloud TPU, a chip designed for machine learning. That's because the AI has been trained, it's "learnt" the data -- it doesn't have to go through it every single time it needs to make a forecast. GenCast improves upon Deepmind's GraphCast model unveiled last year. Other tech firms are also developing their own AI weather forecasters. Nvidia released FourCastNet in 2022, while Huawei launched its Pangu-Weather model in 2023. So will AI replace traditional forecasting soon? Probably not. Models like GenCast still rely on data from traditional weather systems and models to train and calibrate their predictions. However, AI can certainly enhance current methods. "The greatest value comes from a hybrid approach, combining human assessment, traditional physics-based models and AI-based weather forecasting," Steven Ramsdale, chief forecaster at the UK's Met Office, told the Financial Times.

DeepMind AI predicts weather more accurately than existing forecasts

The latest weather forecasting AI model from Google DeepMind can beat the leading providers more than 97 per cent of the time, and it is quicker and cheaper to run Google DeepMind claims its latest weather forecasting AI can make predictions faster and more accurately than existing physics-based simulations. GenCast is the latest in DeepMind's ongoing research project to use artificial intelligence to improve weather forecasting. The model was trained on four decades of historical data from the European Centre for Medium-Range Weather Forecasts's (ECMWF) ERA5 archive, which includes regular measurements of temperature, wind speed and pressure at various altitudes around the globe. Data up to 2018 was used to train the model and then data from 2019 was used to test its predictions against known weather. The company found that it beat ECMWF's industry-standard ENS forecast 97.4 per cent of the time in total, and 99.8 per cent of the time when looking ahead more than 36 hours. Last year, DeepMind worked with ECMWF to create an AI that beat the "gold-standard" high-resolution HRES 10-day forecast more than 90 per cent of the time. Prior to that, it had developed "nowcasting" models that predicted the chance of rain in a given 1-square-kilometre area from 5 to 90 minutes ahead using 5 minutes of radar data. And Google is also working on ways of using AI to replace small parts of deterministic models to speed up computation while retaining accuracy. Existing weather forecasts are based on physics simulations run on powerful supercomputers that deterministically model and extrapolate weather patterns as accurately as possible. Forecasters usually run dozens of simulations with slightly different inputs in groups called ensembles to better capture a range of possible outcomes. These increasingly complex and numerous simulations are extremely computationally intensive and require ever more powerful and energy-hungry machines to operate. AI could offer a less costly solution. For instance, GenCast creates forecasts with an ensemble of 50 possible futures, each taking just 8 minutes on a custom-made and AI-focused Google Cloud TPU v5 chip. GenCast operates with a resolution of cells around 28 square kilometres at the equator. Since the data used in this research was collected, ECMWF's ENS has been upgraded to a resolution of just 9 kilometres. Ilan Price at DeepMind says the AI may not need to follow suit and could offer a way forward without collecting finer data and running more intensive calculations. "When you have a traditional physics-based model, that is a necessary requirement for getting more accurate predictions, because it's a necessary requirement of more accurately solving the physical equations," says Price. "[With] machine learning, [it] isn't necessarily the case that going to higher resolution is a requirement for getting more accurate simulations or predictions out of your model." David Schultz at the University of Manchester, UK, says AI models present an opportunity to make weather forecasts more efficient but they are often overhyped, and it is important to remember that they rely heavily on training data from traditional physics-based models. "Is it [GenCast] going to revolutionise numerical weather prediction? No, because you still have to run the numerical weather prediction models in the first place to train the models," says Schultz. "If you never had ECMWF in the first place, creating the ERA5 reanalyses, and all the investment that went into that, you wouldn't have these AI tools. That's like saying 'I can beat Garry Kasparov at chess, but only after I study every move he ever played'." Sergey Frolov at the US National Oceanic and Atmospheric Administration (NOAA) thinks the AI will need training data with higher resolution to progress further. "What we're fundamentally seeing is that all these approaches are getting stopped [from advancing] by the fidelity of training data," he says. "And the training data comes from operational centres like ECMWF and NOAA. To move this field forward, we need to generate more training data with physics-based models of higher fidelity." But for now, GenCast does offer a way to run forecasts at lower computation cost, and more quickly. Kieran Hunt at the University of Reading, UK, says just as a collection of physics-based forecasts can generate better results than a single forecast, he believes ensembles will boost the accuracy of AI forecasts. Hunt points to the record 40°C (104°F) temperatures seen in the UK in 2022 as an example. A week or two earlier, there were lone members of ensembles predicting it, but they were considered anomalous. Then, as we drew nearer to the heatwave, more and more forecasts fell in line, allowing early warning that something unusual was coming. "It does allow you to hedge a little if there is one member that shows something really extreme; it might happen, but it probably won't," says Hunt. "I wouldn't view it as necessarily a step change. It's combining the tools that we've been using in weather forecasting for a while with the new AI approach in a way that will certainly work to improve the quality of AI weather forecasts. I've no doubt this will do better than the kind of first wave of AI weather forecasts."

Goodbye, unreliable weather forecasts? Google DeepMind's AI model sets new benchmark for 15-day predictions

If you still find yourself getting caught out and frustrated by inaccurate weather forecasts, Google has some good news: DeepMind, its AI research lab, has unveiled a generative model which can forecast the weather with incredible accuracy, up to 15 days in advance. In simple terms, GenCast is an AI model that's similar to ChatGPT. The difference with GenCast is that it's been specifically adapted to the Earth's geometry. Fed with recent weather data, it's able to generate future weather scenarios and suggest the most probable outcome. The team at DeepMind trained GenCast on archival weather data collected between 1979 and 2018. This included temperature, wind speed and air pressure readings from around the globe. Based on this, the model was able to learn global weather patterns. To test its effectiveness, Google then compared the model's predictions against the industry's current best forecasting tool, the Ensemble Forecast (ENS). The model was asked to generate a forecast for 2019, with 1,320 combinations of variables and lead times. GenCast was more accurate than ENS 97.2% of the time. For forecasts more than 36 hours in advance, that increased to 99.8%. That accuracy also applied to the prediction of extreme weather events, such as tropical cyclones. What's even more remarkable is that, according to DeepMind, it took a single Google Cloud Tensor Processing Unit v5 - the circuits that Google uses to accelerate machine learning tasks - only eight minutes to produce a 15-day forecast. Traditional ensemble forecasts take several hours to complete, running on supercomputers with thousands of processors. No meteorologist replacement Traditional ensemble forecasts project weather patterns based on complex physics-based calculations using gathered data. GenCast takes that same data and generates the most likely scenario, informed by everything it's learned from historic weather data. GenCast won't put meteorologists out of work. The model is only as reliable as the data it's trained on and, in a changing climate, past weather patterns may not prove a reliable basis for forecasting far into the future. There are also a number of atmospheric variables which GenCast can't account for, meaning manual equations will still be required to arrive at a dependable prediction. It's likely that GenCast will become another tool in the meteorological arsenal, with its input considered alongside other data sources. Human experts will still be required to make an assessment of the most likely scenario. GenCast also isn't the first use of deep learning to achieve more precise weather forecasts. FourCastNet is a data-driven forecasting model from Nvidia, while Huawei's Pangu-Weather model has proven more accurate than numerical weather prediction in tests. Atmo, a San Francisco company, is also developing AI models to improve forecasting based on real-time weather data. DeepMind sees applications for GenCast beyond standard forecasting. By providing the probability of different weather events, GenCast allows officials to plan for different outcomes and allocate resources accordingly. Google also sees a future where wind-power forecasting could influence renewable energy planning. For now, the team at DeepMind is continuing research and develop GenCast as one of several AI-based weather models, including its deterministic medium-range forecasts. The model is open and DeepMind will soon be releasing real-time and historical forecasts, for integration into other models. You might also like...

What to Know About Google's Breakthrough Weather Prediction Model

The latest forecast tool out of DeepMind shows how artificial intelligence could revolutionize the way we predict the weather. The Sun'll come out tomorrow, and you no longer have to bet your bottom dollar to be sure of it. Google's DeepMind team released its latest weather prediction model this week, which outperforms a leading traditional weather prediction model across the vast majority of tests put before it. The generative AI model is dubbed GenCast, and it is a diffusion model like those undergirding popular AI tools including Midjourney, DALL·E 3, and Stable Diffusion. Based on the team's tests, GenCast is better at predicting extreme weather, the movement of tropical storms, and the force of wind gusts across Earth's mighty sweeps of land. The team's discussion of GenCast's performance was published this week in Nature. Where GenCast departs from other diffusion models is that it (obviously) is weather-focused, and "adapted to the spherical geometry of the Earth," as described by a couple of the paper's co-authors in a DeepMind blog post. Instead of a written prompt such as "paint a picture of a dachshund in the style of Salvador Dalí," GenCast's input is the most recent state of the weather, which the model then uses to generate a probability distribution of future weather scenarios. Traditional weather prediction models like ENS, the leading model from the European Center for Medium-Range Weather Forecasts, make their forecasts by solving physics equations. "One limitation of these traditional models is that the equations they solve are only approximations of the atmospheric dynamics," said Ilan Price, a senior research scientist at Google DeepMind and lead author of the team's latest findings, in an email to Gizmodo. The first seeds of GenCast were planted in 2022, but the model published this week includes architectural changes and an improved diffusion setup that made the model better trained to predict weather on Earth, including extreme weather events, up to 15 days out. "GenCast is not limited to learning dynamics/patterns that are known exactly and can be written down in an equation," Price added. "Instead it has the opportunity to learn more complex relationships and dynamics directly from the data, and this allows GenCast to outperform traditional models." Google has been tooling around with weather prediction for a while, and in recent years have made a couple substantive steps towards more precise forecasting using AI methods. Last year, DeepMind scientistsâ€"some of whom co-authored the new paperâ€"released GraphCast, a machine learning-based method that outperformed the current medium-range weather prediction models on 90% of the targets used in testing. Just five months ago, a team mostly consisting of DeepMind researchers published NeuralGCM, a hybrid weather prediction model that combined a traditional physics-based weather predictor with machine-learning components. That team found that "end-to-end deep learning is compatible with tasks performed by conventional [models] and can enhance the large-scale physical simulations that are essential for understanding and predicting the Earth system." The resolution achieved by GenCast is roughly six times that of NeuralGCM, but that was expected. "NeuralGCM is designed as a general purpose atmospheric model primarily to support climate modelling, whereas the higher resolution of GenCast is often expected for operational medium range forecast models, which is GenCast’s specific target use-case," Price added. "This is also why we emphasized a wide range of evaluations which are crucial use cases for operational medium range forecasts, like predicting extreme weather." In the recent work, the team trained GenCast on historical weather data through 2018, and then tested the model's ability to predict weather patterns in 2019. GenCast outperformed ENS on 97.2% of targets using different weather variables, with varying lead times before the weather event; with lead times greater than 36 hours, GenCast was more accurate than ENS on 99.8% of targets. The team also tested GenCast's ability to forecast the track of a tropical cycloneâ€"specifically Typhoon Hagibis, the costliest tropical cyclone of 2019, which hit Japan that October. GenCast's predictions were highly uncertain with seven days of lead time, but became more accurate at shorter lead times. As extreme weather generates wetter, heavier rainfall, and hurricanes break records for how quickly they intensify and how early in the season they form, accurate prediction of storm paths will be crucial in mitigating their fiscal and human costs. But that's not all. In a proof-of-principle experiment described in the research, the DeepMind team found that GenCast was more accurate than ENS in predicting the total wind power generated by groups of over 5,000 wind farms in the Global Power Plant Database. GenCast's predictions were about 20% better than ENS' with lead times of two days or less, and retained statistically significant improvements up to a week. In other words, the model does not just have value in mitigating disasterâ€"it could inform where and how we deploy energy infrastructure. What does all of this mean for you, O casual appreciator of climate? Well, the DeepMind team has made the GenCast code open source and the models available for non-commercial use, so you can tool around if you're curious. The team is also working on releasing an archive of historical and current weather forecasts. "This will enable the wider research and meteorological community to engage with, test, run, and build on our work, accelerating further advances in the field," Price said. "We have finetuned versions of GenCast to be able to take operational inputs, and so the model could start to be incorporated in operational setting." There is not yet a timeline on when GenCast and other models will be operational, though the DeepMind blog noted that the models are "starting to power user experiences on Google Search and Maps." Whether you're here for the weather or the AI applications, there's plenty to like about GenCast and the broader suite of DeepMind forecasting models. The accuracy of such tools will be paramount for predicting extreme weather events with enough lead time to protect those in harm's way, be it from floods in Appalachia or tornadoes in Florida.

DeepMind AI weather forecaster beats world-class system

Google DeepMind has developed the first artificial intelligence (AI) model of its kind to predict the weather more accurately than the best system currently in use. The model generates forecasts up to 15 days in advance -- and it does so in minutes, rather than the hours needed by today's forecasting programs. The purely AI system beats the world's best medium-range operational model, the European Centre for Medium-Range Weather Forecasts' ensemble model (ENS), at predicting extreme weather such as hurricanes and heatwaves. The breakthrough could help usher in an era of AI weather forecasting that is quicker and more reliable than today's systems, researchers say. The system, called GenCast, is described today in Nature. Conventional forecasts, including those from ENS, are based on mathematical models that simulate the laws of physics governing Earth's atmosphere. They use supercomputers to crunch data from satellites and weather stations -- a process that takes hours and vast amounts of computing power. GenCast, by contrast, has been trained only on historical weather data, which enables the system to draw out complex relationships between variables such as air pressure, humidity, temperature and wind. This helps it to outperform strictly physics-based systems, says Ilan Price, a research scientist at Google DeepMind in London and an author of the paper. "We've really made dramatic progress to catch up and now overtake [physics-based models] with machine learning", Price says. AI weather forecasting has advanced rapidly, with multiple companies racing to develop new and better models. Among them are Huawei in Shenzhen, China, and Nvidia in Santa Clara, California. Earlier this year, Google released NeuralGCM, a hybrid system that combines physics-based models with AI to produce short- and long-term forecasts on a par with conventional models. Some of the AI systems released to date are 'deterministic' models, meaning that they offer only a single forecast and do not estimate the likelihood that the forecast will be correct. By contrast, GenCast generates 'ensemble' forecasts: a suite of forecasts that have each been produced from slightly different starting conditions. By combining these forecasts into an ensemble, scientists can produce a final forecast and estimate the probability that the forecasted weather will occur. Price and his colleagues trained the AI on global weather data from 1979 to 2018 and then predicted the weather of 2019. To check its accuracy, they compared GenCast forecasts with actual weather data and ENS forecasts for that year. GenCast was more accurate than ENS on 97% of the measures used on a scorecard to evaluate such 'probabilistic' forecasts. It was also better at forecasting extreme heat, cold and wind, as well as tropical-cyclone tracks. GenCast produces one 15-day forecast within 8 minutes on an AI processing chip. This speed is "quite substantially faster" than the time it takes conventional models, Price says. The researchers have released the underlying code and are making model parameters called 'weights' available for non-commercial use. Price says this will help to "democratize" research and increase public access to weather modelling. "This is a really great contribution to open science," says Matthew Chantry, a machine-learning coordinator at the European Centre For Medium-Range Weather Forecasts in Reading, UK. "We need to understand how these models perform in the most extreme weather events", and publishing the model and data publicly will allow the research community to assess them, he says. Chantry read a manuscript of the paper when it was posted on a preprint archive last year and was inspired by GenCast's 'diffusion' approach, which introduces random noise into the model to refine its reliability. "We've actually implemented some of the key breakthroughs in our own machine-learning model," he says. The resulting model, called Artificial Intelligence/Integrated Forecasting System (AIFS), will be published soon, he adds. Having more accurate forecasts sooner can help people make informed decisions, Price says, especially for those living in the path of a hurricane.

Google's AI can accurately predict weather forecasts 15 days in...

"Googling" today's forecast could soon be even more reliable. Just like your friendly TV meteorologist, current weather models could perhaps be a thing of the past. Google has unveiled an AI meteorology tech that is far faster and more accurate than traditional forecasts, per a study published in the journal "Nature." Devised by the search engine firm's AI division, DeepMind, the "GenCast" model can tell if it's going to rain 15 days ahead of time at a higher accuracy rate than the European Centre for Medium-Range Weather Forecasts' ENS (ECMWF) -- the world's top operational forecasting system, per the Google Deepmind blog. This discrepancy has to do with a completely new monsoon-divining methodology. Whereas current iterations are "deterministic, and provided a single, best estimate of future weather," GenCast "comprises an ensemble of 50 or more predictions, each representing a possible weather trajectory," the blog's authors write. This vast spread of data allows forecasts to predict the weather with far greater precision, specifically between 97.2% and 99.8% accuracies, depending on circumstances. "Such ensemble forecasts are more useful than relying on a single forecast, as they provide decision-makers with a fuller picture of possible weather conditions in the coming days and weeks and how likely each scenario is," reads the blog. GenCast also employs a type of AI called a diffusion model that's typically found in video, image and music generators. Unlike most versions, this cybernetic weatherperson has been adapted to the spherical geometry of the Earth and is trained on four decades of historical weather data (leading up to 2018) from the ECMWF ranging from temperature to pressure at various altitudes. To evaluate the typhoon-detector's efficacy, researchers compared GenCast's predictions to the real weather data from 2019 and the ENS forecasts for that year. They specifically looked at 1,320 combinations of different variables at different lead times. The AI acid test revealed that the GenCast was more precise than ENS on 97.2% of these targets, and on 99.8% when the lead times were greater than 36 hours. This veritable Deep Blue of weather detection was also far more efficient. It reportedly takes a single Google Cloud processor just eight minutes to create one 15-day forecast in GenCast's ensemble, compared to the hours required to generate physics-based ensemble forecasts -- such as those produced by ENS -- on a supercomputer with tens of thousands of processors. In addition, GenCast could better forecast extreme weather events -- extreme heat and cold, and high wind speeds -- which could help meteorologists keep better tabs on hurricanes and typhoons. The only downside is that the current ENS system can generate significantly higher-resolution forecasts than its AI counterpart, the Daily Mail reported. DeepMind reps also admitted that the current meteorology machines are irreplaceable for now because, for one, they provide the data used to train models like GenCast.

GenCast: New AI model is redefining weather forecasting - Earth.com

Weather shapes our daily lives, influencing decisions about safety, travel, and work. As climate change drives more extreme events, accurate forecasts become indispensable. However, predicting weather patterns beyond the next few days remains challenging. Probabilistic ensemble forecasts have emerged as an essential prediction tool, offering a range of possible scenarios to inform decisions. Now, a new AI model, GenCast, aims to redefine weather forecasting with unmatched precision and speed. Developed by researchers and recently featured in the journal Nature, GenCast is a high-resolution AI ensemble model that is capable of forecasting weather for up to 15 days ahead. Compared to the European Centre for Medium-Range Weather Forecasts' (ECMWF) ENS system, GenCast consistently delivers more accurate predictions. The model is not only faster but also highly reliable in predicting both daily weather patterns and extreme events like cyclones and heat waves. GenCast represents a significant leap from its predecessor - a deterministic model that offered a single best estimate. Instead, GenCast uses an ensemble of 50 or more predictions, each representing a potential weather trajectory. This approach provides a nuanced view of possible outcomes that is crucial for effective decision-making. GenCast utilizes diffusion models, a type of generative AI previously used in creating images, videos, and music. Adapted for Earth's spherical geometry, the model accurately simulates complex weather patterns. Researchers trained GenCast using four decades of historical data from ECMWF's ERA5 archive, which includes variables such as temperature, wind speed, and pressure at multiple altitudes. The result is a model with an unparalleled understanding of global weather dynamics. The researchers evaluated GenCast using data from 2019, testing it on 1,320 combinations of variables and lead times. GenCast outperformed the ENS in 97.2% of these cases and excelled in 99.8% of scenarios at lead times beyond 36 hours. These results demonstrate GenCast's superior skill in predicting both standard weather conditions and extreme events. For example, in forecasting extreme heat or high winds, GenCast consistently provided higher economic value. Its ensemble forecasts balance confidence and uncertainty, offering actionable insights for scenarios like cyclone paths. One striking example was Typhoon Hagibis. Seven days before landfall, GenCast forecasted a wide range of possible paths, which narrowed into a precise and accurate cluster as the storm approached Japan. GenCast's efficiency is another standout feature. While traditional models require hours on supercomputers, GenCast generates a 15-day forecast in just eight minutes using a single Google Cloud TPU v5. This speed and scalability make it an ideal tool for real-time applications and broader accessibility. The potential applications of GenCast extend far beyond weather forecasting. Accurate predictions of extreme weather risks can save lives, prevent damage, and reduce costs. For instance, improved forecasts of tropical cyclones enable better disaster preparedness, while enhanced wind power predictions can boost renewable energy reliability. GenCast's precision in forecasting renewable energy outputs gives it immense societal value. Google emphasizes collaboration and transparency by making GenCast's code and weights openly available. This initiative aims to accelerate research and innovation in weather forecasting and climate science. The model's real-time and historical forecasts will soon be accessible to researchers, meteorologists, and industries. The AI model complements Google's suite of AI-based weather models, including NeuralGCM, SEEDS, and flood prediction systems. These models already enhance user experiences on platforms like Google Search and Maps. By combining AI and traditional meteorology, Google aims to refine forecasts and address challenges posed by climate change. Looking ahead, GenCast's impact could extend to sectors like food security and disaster response. Partnerships with renewable energy companies and organizations focused on humanitarian efforts offer opportunities for significant societal benefits. As GenCast continues to evolve, it exemplifies how AI can bridge the gap between data and actionable insights, thereby addressing critical challenges in weather prediction and beyond. GenCast stands at the forefront of a new era in weather forecasting that merges advanced AI with traditional methods to serve society better in a changing climate. Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates.

Google DeepMind predicts weather more accurately than leading system

AI program GenCast performed better than ENS forecast at predicting day-to-day weather and paths of hurricanes and cyclones For those who keep an eye on the elements, the outlook is bright: researchers have built an artificial intelligence-based weather forecast that makes faster and more accurate predictions than the best system available today. GenCast, an AI weather program from Google DeepMind, performed up to 20% better than the ENS forecast from the European Centre for Medium-Range Weather Forecasts (ECMWF), widely regarded as the world leader. In the near term, GenCast is expected to support traditional forecasts rather than replace them, but even in an assistive capacity it could provide clarity around future cold blasts, heatwaves and high winds, and help energy companies predict how much power they will generate from windfarms. In a head-to-head comparison, the program churned out more accurate forecasts than ENS on day-to-day weather and extreme events up to 15 days in advance, and was better at predicting the paths of destructive hurricanes and other tropical cyclones, including where they would make landfall. "Outperforming ENS marks something of an inflection point in the advance of AI for weather prediction," said Ilan Price, a research scientist at Google DeepMind. "At least in the short term, these models are going to accompany and be alongside existing, traditional approaches." Traditional physics-based weather forecasts solve vast numbers of equations to produce their predictions, but GenCast learned how global weather evolves by training on 40 years of historic data generated between 1979 and 2018. This included wind speed, temperature, pressure, humidity and dozens more variables at different altitudes. Given the latest weather data, GenCast predicts how conditions will change around the planet in squares of up to 28km by 28km for the next 15 days in 12-hour steps. While a traditional forecast takes hours to run on a supercomputer with tens of thousands of processors, GenCast takes only eight minutes on a single Google Cloud TPU, a chip designed for machine learning. Details are published in Nature. Google has released a string of AI-powered weather forecasts in recent years, the fruits of researchers dabbling with different approaches. In July, the firm announced NeuralGCM, which combines AI and traditional physics for long range forecasts and climate modelling. In 2023, Google DeepMind unveiled GraphCast, which produces one single best-guess forecast at a time. GenCast builds on GraphCast by generating an ensemble of 50 or more forecasts, assigning probabilities for different weather events ahead. Weather forecasters welcomed the advance. Steven Ramsdale, a Met Office chief forecaster with responsibility for AI, said the work was "exciting", while a spokesperson for the ECMWF called it "a significant advance", adding that components of GenCast were being used in one of its AI forecasts. "Weather forecasting is on the brink of a fundamental shift in methodology," said Sarah Dance, professor of data assimilation at the University of Reading. "This opens up the possibility for national weather services to produce much larger ensembles of forecasts, providing more reliable estimates of forecast confidence, particularly for extreme events." But questions remain. "The authors have not answered whether their system has the physical realism to capture the 'butterfly effect', the cascade of fast-growing uncertainties, which is critical for effective ensemble forecasting," Prof Dance said. "There is still a long way to go before machine learning approaches can completely replace physics-based forecasting," she added. The data GenCast trained on combines past observations with physics-based "hindcasts" that need sophisticated maths to fill gaps in historic data, she said. "It remains to be seen whether generative machine learning can replace this step and go straight from the most recent unprocessed observations to a 15-day forecast," Dance said. The performance is promising, but is a "Michael Fish moment" lurking on the horizon? "Will AI forecasting be immune?" said Price. "All prediction models would have the chance of making an error and GenCast is no different."

Google unveils new AI weather model that beats traditional forecasts

While things like generating podcasts and videos are nice, we've long hoped our newfound grasp of artificial intelligence might spur on some more genuinely useful improvements that mankind can benefit from. Thankfully, we've had examples of this with the detection of 160,000 new viruses, but a new boast from Google might make a difference for just about everyone. The tech giant's DeepMind division focused on pushing the limits of AI development, has released a blog post describing how its new GenCast model can predict weather and extreme conditions more accurately than leading tools. In fact, it can deliver "faster, more accurate forecasts up to 15 days ahead", which might mean your local weather presenter's days might be numbered. The current leader in weather forecasting, ENS, is maintained by the European Centre for Medium-Range Weather Forecasts (ECMWF - they really need better branding), but in testing, Google's new GenCast model performed up to 20% faster. It's not expected to replace current weather forecasting technology, but could supplement it in the near future, but it's perhaps most useful in more extreme conditions. According to Google, it could help anticipate hazards like heatwaves and cold blasts, and it beat ENS when it came to predicting the paths of hurricanes and cyclones.

AI weather models can now beat the best traditional forecasts

A new machine-learning weather prediction model called GenCast can outperform the best traditional forecasting systems in at least some situations, according to a paper by Google DeepMind researchers published today in Nature. Using a diffusion model approach similar to artificial intelligence (AI) image generators, the system generates multiple forecasts to capture the complex behavior of the atmosphere. It does so with a fraction of the time and computing resources required for traditional approaches. How weather forecasts work The weather predictions we use in practice are produced by running multiple numerical simulations of the atmosphere. Each simulation starts from a slightly different estimate of the current weather. This is because we don't know exactly what the weather is at this instant everywhere in the world. To know that, we would need sensor measurements everywhere. These numerical simulations use a model of the world's atmosphere divided into a grid of three-dimensional blocks. By solving equations describing the fundamental physical laws of nature, the simulations predict what will happen in the atmosphere. Known as general circulation models, these simulations need a lot of computing power. They are usually run at high-performance supercomputing facilities. Machine-learning the weather The past few years have seen an explosion in efforts to produce weather prediction models using machine learning. Typically, these approaches don't incorporate our knowledge of the laws of nature the way general circulation models do. Most of these models use some form of neural network to learn patterns in historical data and produce a single future forecast. However, this approach produces predictions that lose detail as they progress into the future, gradually becoming "smoother." This smoothness is not what we see in real weather systems. Researchers at Google's DeepMind AI research lab have just published a paper in Nature describing their latest machine-learning model, GenCast. GenCast mitigates this smoothing effect by generating an ensemble of multiple forecasts. Each individual forecast is less smooth, and better resembles the complexity observed in nature. The best estimate of the actual future then comes from averaging the different forecasts. The size of the differences between the individual forecasts indicates how much uncertainty there is. According to the GenCast paper, this probabilistic approach creates more accurate forecasts than the best numerical weather prediction system in the world -- the one at the European Center for Medium-Range Weather Forecasts. Generative AI -- for weather GenCast is trained on what is called reanalysis data from the years 1979 to 2018. This data is produced by the kind of general circulation models we talked about earlier, which are additionally corrected to resemble actual historical weather observations to produce a more consistent picture of the world's weather. The GenCast model makes predictions of several variables such as temperature, pressure, humidity and wind speed at the surface and at 13 different heights, on a grid that divides the world up into 0.25-degree regions of latitude and longitude. GenCast is what is called a "diffusion model," similar to AI image generators. However, instead of taking text and producing an image, it takes the current state of the atmosphere and produces an estimate of what it will be like in 12 hours. This works by first setting the values of the atmospheric variables 12 hours into the future as random noise. GenCast then uses a neural network to find structures in the noise that are compatible with the current and previous weather variables. An ensemble of multiple forecasts can be generated by starting with different random noise. Forecasts are run out to 15 days, taking 8 minutes on a single processor called a tensor processor unit (TPU). This is significantly faster than a general circulation model. The training of the model took five days using 32 TPUs. Machine-learning forecasts could become more widespread in the coming years as they become more efficient and reliable. However, classical numerical weather prediction and reanalyzed data will still be required. Not only are they needed to provide the initial conditions for the machine learning weather forecasts, they also produce the input data to continually fine-tune the machine learning models. What about the climate? Current machine learning weather forecasting systems are not appropriate for climate projections, for three reasons. Firstly, to make weather predictions weeks into the future, you can assume that the ocean, land and sea ice won't change. This is not the case for climate predictions over multiple decades. Secondly, weather prediction is highly dependent on the details of the current weather. However, climate projections are concerned with the statistics of the climate decades into the future, for which today's weather is irrelevant. Future carbon emissions are the greater determinant of the future state of the climate. Thirdly, weather prediction is a "big data" problem. There are vast amounts of relevant observational data, which is what you need to train a complex machine learning model. Climate projection is a "small data" problem, with relatively little available data. This is because the relevant physical phenomena (such as sea levels or climate drivers such as the El Niño-Southern Oscillation) evolve much more slowly than the weather. There are ways to address these problems. One approach is to use our knowledge of physics to simplify our models, meaning they require less data for machine learning. Another approach is to use physics-informed neural networks to try to fit the data and also satisfy the laws of nature. A third is to use physics to set "ground rules" for a system, then use machine learning to determine the specific model parameters. Machine learning has a role to play in the future of both weather forecasting and climate projections. However, fundamental physics -- fluid mechanics and thermodynamics -- will continue to play a crucial role.

Google DeepMind's GenCast model is ready to upend traditional weather forecasts

Google's new AI weather summary can be useful when you're in a rush Key Takeaways Google DeepMind's new ML model, GenCast, can accurately predict weather patterns up to 15 days in advance. GenCast has been found to surpasses traditional weather forecasting methods, including the European Center for Medium-Range Weather Forecasts (ECMWF). GenCast offers probabilistic forecasts, enhancing users' understanding of multiple weather scenarios. The model has the potential to significantly improve weather forecasting and its applications in various sectors. ✕ Remove Ads Traditional weather forecast models fail to reliably predict weather patterns beyond a week, but a new tool from Google's AI subsidiary, DeepMind, seems to be breaking the limitation. GenCast, as it has been titled, is currently outperforming the world's leading weather forecasting systems, including the European Center for Medium-Range Weather Forecasts (ECMWF), with evidence pointing to its ability to generate accurate 15-day forecasts in minutes. Related The 10 best weather apps on Android in 2024 Never get caught without your umbrella Posts1 The findings were first published in the journal Nature (via The New York Times), and it highlights how Machine Learning (ML) based weather prediction has less forecast error than traditional numerical weather prediction (NWP), "which relies on physics-based simulations of the atmosphere." GenCast, for reference, was trained on decades of reanalysis data, allowing it to generate 15-day global forecasts in just eight minutes. Supercomputers used for ECMWF, on the other hand, take hours. ✕ Remove Ads Researchers pinned GenCast against the traditional ECMWF, which, for reference, is used by 35 nations for their official weather forecast needs. In the context of determining a designated set of 1,320 global wind speeds, temperatures and other atmospheric features, GenCast "outdid the center's forecasts 97.2 percent of time." A leap beyond last year's GraphCast Related Google's weather forecasting tool is already blowing away conventional models GraphCast uses AI to forecast up to 10 days of weather in less than 1 minute Posts It's worth noting that this isn't DeepMind's first foray into weather forecasting. Last year, the tech giant debuted GraphCast, which was able to accurately predict weather up to 10 days in advance. One of GenCast's key advancements, however, lies in its probabilistic forecasting capabilities. This means that while GraphCast provides a single, definitive forecast, GenCast can offer a range of potential outcomes -- which can better help users understand and prepare for the likelihood of different weather scenarios. For example, in GraphCast's case, it might say "rain expected tomorrow," while GenCast can offer a probabilistic forecast of "70 percent chance of rain tomorrow." ✕ Remove Ads On a more user-focused front, Google's mobile weather app uses a related model called NowCast. Given the impressive capabilities of GenCast, it is highly plausible that the ML weather forecasting tool might one day be able to power the Pixel weather app, providing users with 15-day forecasts that they can actually rely on. However, Kerry Emanuel, a professor emeritus of atmospheric science at MIT suggests that GenCast won't replace traditional forecasting methods, but it sure can complement them. For now, GenCast's weather predictions are expected to make their way to the Google Earth Engine and BigQuery, giving scientists and researchers worldwide access to analyze DeepMind's forecasts. Related Google's new AI weather summary can be useful when you're in a rush Get weather info at a glance Posts3 ✕ Remove Ads

Google DeepMind hits new milestone in AI weather forecasting

Google DeepMind has unveiled an artificial intelligence weather prediction model that outperforms traditional methods on forecasts up to 15 days and is better at foreseeing extreme events. The tool, known as GenCast, gauges the likelihood of multiple scenarios to accurately estimate trends from wind power production to tropical cyclone movements. GenCast's probabilistic technique is a new milestone in the rapid progress in using AI to power better and quicker everyday weather projections, an approach big traditional forecasters are increasingly embracing. "[This] marks something of an inflection point in the advance of AI for weather prediction, with state of the art raw forecasts now coming from machine-learning models," said Ilan Price, a research scientist at Google DeepMind. "GenCast could be incorporated as part of operational weather forecasting systems, offering valuable insights to help decision makers better understand and prepare for upcoming weather events." GenCast's novelty over previous machine-learning models is its use of the so-called "ensemble" predictions representing different outcomes, a technique deployed in state-of-the-art traditional forecasting. GenCast is trained on four decades of data from the European Centre for Medium-Range Weather Forecasting (ECMWF). The model outperformed the ECMWF's 15-day forecast on 97.2 per cent of 1,320 variables, such as temperature, wind speed and humidity, according to a paper published in Nature on Wednesday. The results are a further improvement in accuracy and scope on Google DeepMind's breakthrough GraphCast model unveiled last year. GraphCast outperformed the ECMWF's forecasts on about 90 per cent of metrics for predictions three to 10 days ahead. AI forecasting models are typically faster and potentially more efficient than standard forecasting methods, which rely on vast computing power to crunch equations derived from atmospheric physics. GenCast can generate its prediction in just eight minutes, compared with hours for the traditional forecast -- and with a fraction of the electronic processing needs. The GenCast model could be further improved in areas such as its ability to predict the intensity of big storms, the researchers said. The resolution of its data could be increased to match upgrades made this year by the ECMWF. The ECMWF said the development of GenCast was a "significant milestone in the evolution of weather forecasting". It said it had integrated "key components" of the GenCast approach in a version of its own AI forecasting system, with live ensemble predictions available since June. The innovative machine-learning science behind GenCast still needed to be tested on extreme weather events, the ECMWF added. The development of GenCast will further fuel debate about how extensively AI should be deployed in forecasting, with many scientists preferring a hybrid technique for some purposes. In July, Google unveiled the NeuralGCM model, which combines machine learning and traditional physics to achieve better results than AI alone for long-range forecasting and climate trends. "There are open questions and discussion about the optimal balance between physics and machine learning forecasting systems. A wide scientific community including [us] is actively exploring this", the ECMWF said. The UK Met Office, the national weather service, is researching how to harness the "exciting" developments to its own AI-driven forecasting models, said Steven Ramsdale, chief forecaster with responsibility for AI. "We maintain the greatest value comes from a hybrid approach, combining human assessment, traditional physics-based models and AI-based weather forecasting," he added.

Google's New AI Weather Model Nails 15-Day Forecasts

Samantha Kelly is a freelance writer with a focus on consumer technology, AI, social media, Big Tech, emerging trends and how they impact our everyday lives. Her work has been featured on CNN, NBC, NPR, the BBC, Mashable and more. Google's new AI weather model could redefine how we predict and prepare for changing conditions. Google DeepMind, the company's AI research division focused on solving scientific problems, announced a new AI weather model called GenCast, which promises to provide more accurate probabilities of different weather conditions up to 15 days in advance. Although weather forecasting typically involves physics-based models, which can take hours to compute on massive supercomputers, Google said it aims to achieve greater accuracy in only minutes. GenCast can generate a 15-day forecast scenario in just 8 minutes using a Tensor Processing Unit chip, according to the company. GenCast is part of Google's expanding AI-powered weather model suite, which includes enhancing Google Search and Maps with improved forecasting for precipitation, wildfires, flooding and extreme heat. Details of its latest testing were published this week in the journal "Nature." Although Google's AI model provided one best estimate of future weather, GenCast uses probability-based forecasting with 50 or more predictions of how the weather may change and assessing the likelihood of those scenarios. The technology runs on a diffusion model, similar to the machine learning models used in generative AI. "We trained it on 40 years of historical data from [the European Centre for Medium-Range Weather Forecasts], which included variables such as temperature, wind speed, and pressure at various altitudes -- enabling it to learn," the company explained in a tweet. When predicting extreme heat, cold and high wind speeds, Google said GenCast outperformed current forecasting models. It also noted that it delivers "superior predictions" of the path of tropical cyclones up to five days in advance.

Google DeepMind unveils GenCast, an AI tool improving 15-day weather for

GenCast was, on average, more accurate than the European Centre for Medium-Range Weather Forecasts (ECMWF). Recently, Google DeepMind announced GenCast, a new artificial intelligence weather prediction tool capable of generating high-resolution forecasts up to 15 days in advance. According to Nature, GenCast significantly improves weather forecasting, marking a major advancement in the field. Unlike traditional weather models that rely on solving physics equations, GenCast is a machine learning-based weather model that learns directly from historical weather data. "GenCast is not limited to learning dynamics/patterns that are known exactly and can be written down in an equation," said Ilan Price, a research scientist at Google DeepMind and an author of the paper published in Nature, according to The Register. GenCast merges computational approaches used by atmospheric scientists with a diffusion model commonly used in generative AI, maintaining high resolution while significantly cutting computational costs, according to the Nature paper. It produces a 15-day weather forecast in just eight minutes, which is a significant improvement over traditional methods that can take hours to generate forecasts. In tests comparing the 15-day forecasts generated for weather in 2019, GenCast was, on average, more accurate than the European Centre for Medium-Range Weather Forecasts (ECMWF) Ensemble Prediction System (ENS) 97.2 percent of the time. With lead times greater than 36 hours, GenCast was 99.8 percent more accurate than ECMWF's ENS system. GenCast consistently outperformed ENS in forecasting extreme weather, including extreme heat, cold, and high wind speeds, as noted by Business Insider. It also demonstrated superior forecasting capability in predicting the path of tropical cyclones, according to The Register. The AI tool helps in better understanding and monitoring dangerous phenomena such as severe storms and hurricanes, allowing necessary protective measures to be taken, which can safeguard lives and reduce damage. "Better predictions of extreme weather enable better decisions," Google's DeepMind said in an announcement, according to the Financial Times. Rémi Lam of DeepMind noted that GenCast's generative skills were rooted in factual data gathered from nature rather than the internet. "We have a ground truth. We have a reality check," Lam said, as reported by The New York Times. Experts in the field have recognized the significance of GenCast's advancements."It's a big deal; it's an important step forward," said Kerry Emanuel, a professor emeritus of atmospheric science at MIT who did not participate in DeepMind's research, according to The New York Times. "The status quo isn't going to disappear. Perhaps the two of them working together will prove to be the best way forward," he added. GenCast's probabilistic forecasting capabilities provide a range of percentages for the likelihood of various weather scenarios, enhancing users' understanding and decision-making in high-risk situations, according to The New York Times. Stay updated with the latest news! Subscribe to The Jerusalem Post Newsletter Subscribe Now The ECMWF acknowledged the advancements made by GenCast. Matthew Chantry, an AI specialist at the ECMWF, confirmed that his agency was already adopting some features of DeepMind's GenCast. "We've actually implemented some of the key breakthroughs in our own machine-learning model," he said. Price said that GenCast's weather predictions would soon be posted publicly on Google's Earth Engine and BigQuery, giving scientists access to the new forecasts, according to Engadget. "We're excited for the community to use and build on our research," Price said, reported The New York Times. Despite its high resolution, DeepMind estimates that a single instance of GenCast can be run out to 15 days on Google's tensor processing systems in just eight minutes, according to Ars Technica. This speed makes GenCast's projections much timelier, providing a significant advantage in monitoring rapidly moving storms. This article was written in collaboration with generative AI company Alchemiq

Your Next Weather Forecaster May Be Powered by AI, and It's Surprisingly Good

I Tried NordVPN's Free Remote Connection Tool and Found It Super Useful If AI's good at one thing, it's churning through lots of data and building a model based on what it "experienced." So, what happens when you feed an AI 39 years' worth of global weather data and then ask it to make a forecast for the future? Well, if Google's analysis is correct, you end up with a system that's more accurate than human measurements. ✕ Remove Ads Google DeepMind's AI Model Beats Human Predictions 97.2% of the Time As reported by the Google DeepMind research lab, the company has had great success using AI to predict the future. The researchers fed their new model, called "GenCast," all of the global weather data between 1979 and 2018. It was then asked to predict what it thought would happen in 2019, comparing its findings to what the Ensemble Forecast predicted and what the weather was actually like on those days. It turns out that AI managed to beat the Ensemble Forecast 97.2% of the time. That comparison rises to 99.8% when GenCast was asked to do forecasts past the 36-hour mark, making it much better than humans for making more distant forecasts. ✕ Remove Ads The most interesting part is how GenCast's accuracy is also applied to predicting extreme weather patterns like tornados. As such, if GenCast behaves in a live situation the same way it did in its tests, it may prove to be an invaluable asset for catching dangerous weather patterns and calling for evacuations far quicker than before. So, is this the end for meteorologists? Probably not. After all, GenCast's accurate forecasts wholly depend on receiving good data. As such, experts will likely use the AI models alongside their current systems to quickly and accurately generate weather forecasts instead of being replaced by them outright. If you want to harness the power of Google's AI, the company has a model called Gemini that's designed as an ideal personal companion. Check out how Gemini can help you be more productive in Google Workspace, or these useful tricks for using Gemini AI in PowerPoint. ✕ Remove Ads

Google's GenCast promises 99.8% accuracy for weather forecasts

Google DeepMind has unveiled GenCast, a groundbreaking AI ensemble model that enhances weather forecasting accuracy and speed significantly. This model addresses the crucial need for reliable forecasts, especially as climate change increases extreme weather occurrences. GenCast predicts a range of possible weather scenarios, outperforming the European Centre for Medium-Range Weather Forecasts' (ECMWF) ENS system. Google DeepMind launches GenCast for enhanced weather forecasting The introduction of GenCast is particularly timely, as the demand for precise weather forecasts continues to grow. The model accurately predicts day-to-day weather changes and extreme conditions up to 15 days in advance. GenCast offers a comprehensive view of potential weather patterns, which is vital for decision-makers in various sectors. GenCast employs a high-resolution format of 0.25°, generating 50 or more predictions for different weather trajectories. This approach allows the model to represent uncertainties more effectively compared to traditional forecasting methods. Weather agencies and scientists rely on ensemble forecasts to understand the range of likely scenarios, a necessity given the inherent unpredictability of weather. Google DeepMind co-founder says: "Huge AI funding leads to hype and grifting" To develop GenCast, researchers utilized four decades of ECMWF's historical weather data, which includes various atmospheric variables crucial for accurate predictions. Consequently, the model has demonstrated superior forecasting skills in extensive evaluations, surpassing ECMWF's ENS in 97.2% of tested targets, and achieving 99.8% accuracy for forecasts over 36 hours ahead. Unlike its predecessor, which provided a single estimated forecast, GenCast employs a diffusion model akin to those used in generative AI for multimedia content generation. This adaptation allows GenCast to operate on the spherical geometry of the Earth, enabling it to grasp and model complex weather scenarios. The computational efficiency of GenCast is noteworthy. A single forecast can be generated in just eight minutes using a Google Cloud TPU v5, while traditional methods require hours and substantial computing resources. This time reduction not only increases operational efficiency but also allows for timely decision-making in critical weather situations. Enhanced predictions for extreme weather events GenCast has excelled in forecasting extreme weather, crucial for public safety and resource management. During testing, the model demonstrated superior abilities in predicting instances of extreme heat, cold, and high wind speeds. For instance, it provided precise tracking of Typhoon Hagibis days before landfall, showcasing its ability to hone in on specific cyclone paths with enhanced accuracy. Furthermore, more reliable weather forecasts can facilitate better planning for renewable energy initiatives. An example includes GenCast's notable accuracy in predicting wind power generation, thereby supporting the transition to sustainable energy sources. This capability is increasingly vital as industries seek dependable data to enhance operational efficiency.

Google DeepMind unveils new AI model that can predict weather

The tech giant claimed that its new AI model can predict both day-to-day weather and extreme events better than the top operational system. Google DeepMind has revealed that its new artificial intelligence (AI) model, GenCast, can help to accurately predict the weather and detect extreme weather risks. In a paper published yesterday (4 December) in Nature, the company presented its new high-resolution AI ensemble model. Google DeepMind has claimed that the new model can provide better forecasts of both day-to-day weather and extreme events than the top operational system, the European Centre for Medium-Range Weather Forecasts' ENS, up to 15 days in advance. The company added that it will release the model's code, weights and forecasts, in order to support the weather forecasting community. "Because a perfect weather forecast is not possible, scientists and weather agencies use probabilistic ensemble forecasts, where the model predicts a range of likely weather scenarios," said Google DeepMind in blogpost. "Such ensemble forecasts are more useful than relying on a single forecast, as they provide decision-makers with a fuller picture of possible weather conditions in the coming days and weeks and how likely each scenario is." Google DeepMind also explained that in order to evaluate GenCast's performance, it trained the model on historical weather data up to 2018 and tested it on data from 2019. "We comprehensively tested both systems, looking at forecasts of different variables at different lead times - 1,320 combinations in total. GenCast was more accurate than ENS on 97.2pc of these targets, and on 99.8pc at lead times greater than 36 hours." Bringing AI to the weather forecasting table Commenting further on the potential benefits of GenCast, Google DeepMind maintained that more accurate forecasts of risks of extreme weather can help officials to protect more lives, avert damage, and save money. "When we tested GenCast's ability to predict extreme heat and cold, and high wind speeds, GenCast consistently outperformed ENS. "Now consider tropical cyclones, also known as hurricanes and typhoons. Getting better and more advanced warnings of where they'll strike land is invaluable." Given the number of dangerous weather events, such as heatwaves and heavy rainfall, over the past few years, it is understandable why AI is being used to improve weather forecasts. The prospect of using AI to predict the weather has even been welcomed by the World Meteorological Organization (WMO). Google DeepMind originated as a research company and was acquired by Google in 2014. It later merged with the tech giant's other research team, Google Brain, in order to spearhead the company's AI efforts. DeepMind first made reference to weather prediction capabilities in November of 2023, when it claimed that its AI model GraphCast could make accurate, fast predictions of the weather and provide earlier warnings of extreme storms. In October, Demis Hassabis, the co-founder and CEO of DeepMind, along with one of its senior research scientists, John M Jumper, made up two members out of a team of three scientists who were jointly awarded the 2024 Nobel Prize for Chemistry. The trio were awarded the prestigious prize for their discoveries in protein design and protein structure prediction. Yesterday, Google DeepMind announced Genie 2, a separate AI model which it claims can generate interactive worlds akin to those seen in modern video games. Don't miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic's digest of need-to-know sci-tech news.

Google's new AI weather model beats most accurate forecast system

Why it matters: The new model provides -- for the first time -- a machine learning-driven method of conducting ensemble-based forecasting, in which the same model is run with many different initial conditions, to generate probability-based projections. Driving the news: Google DeepMind researchers told reporters on Monday that the model shows skill in anticipating extreme events outside the bounds of what occurred during the training period. The big picture: AI-based weather forecasting is about to be further integrated into government weather and climate agencies along with private companies, experts told Axios. Models like the new "GenCast" are also unlikely to fully displace the current generation of weather models. Zoom in: The new model is detailed in a study published in the journal Nature on Wednesday. The intrigue: Today, top forecast centers such as ECMWF in the U.K. and NOAA prepare their forecasts using ensemble methods rooted mainly in physics-based models. The new-generation of AI weather models that are rolling out from tech companies are trained on past weather data and rely on machine learning techniques. Friction point: Some outside meteorologists told Axios that GenCast's output and skill have gaps. Zoom out: GenCast is one of a slew of AI-driven weather models unveiled in recent years from companies such as Nvidia and Microsoft, each of which has shown promise in improving forecast accuracy. What they're saying: "It is exciting that machine-learning will change the game when it comes to probabilistic forecasting," Marshall Shepherd, a research meteorologist at the University of Georgia who didn't take part in the new study, told Axios.

DeepMind's GenCast AI is really good at forecasting the weather

When Helene made landfall in Florida earlier this year, 234 people lost their lives to the worst hurricane to strike the US mainland since Katarina in 2005. It's natural disasters like that, and their growing intensity due to climate change, that have pushed scientists to develop more accurate weather forecasting systems. On Wednesday, Google's DeepMind division announced what may go down as the most significant advancement in the field in nearly eight decades of work. In a post on the Google Keyword blog, DeepMind's Ilan Price and Matthew Wilson detailed GenCast, the company's latest AI agent. According to DeepMind, GenCast is not only better at providing daily and extreme weather forecasts than its previous AI weather program, but it also outperforms the best forecasting system in use right now, one that's maintained by the European Center for Medium-Range Weather Forecasts (ECMWF). In tests comparing the 15-day forecasts the two systems generated for weather in 2019, GenCast was, on average, more accurate than ECMWF's ENS system 97.2 percent of the time. With lead times greater than 36 hours, DeepMind's was an even better 99.8 percent more accurate. "I'm a little bit reluctant to say it, but it's like we've made decades worth of improvements in one year," Rémi Lam, the lead scientist on DeepMind's previous AI weather program, told The New York Times. "We're seeing really, really rapid progress." GenCast is a diffusion model, which is the same tech that powers Google's generative AI tools. DeepMind trained the software on nearly 40 years of high-quality weather data curated by the European Center for Medium-Range Weather Forecasts. The predictions the new model generates are probabilistic, meaning they account for a range of possibilities that are then expressed as percentages. Probabilistic models are considered more nuanced and useful than their deterministic counterparts, which only offer a best guess of what the weather might be like on a given day. The former also harder to create and calculate. Indeed, what's perhaps most striking about GenCast is that it requires significantly less computing power than traditional physics-based ensemble forecasts like ENS. According to Google, a single one of its TPU v5 tensor processing units can produce a 15-day GenCast forecast in eight minutes. By contrast, it can take a supercomputer with tens of thousands of processors hours to produce a physics-based forecast. Of course, GenCast isn't perfect. One area the software could provide better predictions on is hurricane intensity, though the DeepMind team told The Times it was confident it could find solutions for the agent's current shortcomings. In the meantime, Google is making GenCast an open model, with example code for the tool available on GitHub. GenCast predictions will also soon make their way to Google Earth.

Google's DeepMind tackles weather forecasting, with great performance

By some measures, AI systems are now competitive with traditional computing methods for generating weather forecasts. Because their training penalizes errors, however, the forecasts tend to get "blurry" -- as you move further ahead in time, the models make fewer specific predictions since those are more likely to be wrong. As a result, you start to see things like storm tracks broadening and the storms themselves losing clearly defined edges. But using AI is still extremely tempting because the alternative is a computational atmospheric circulation model, which is extremely compute-intensive. Still, it's highly successful, with the ensemble model from the European Centre for Medium-Range Weather Forecasts considered the best in class. In a paper being released today, Google's DeepMind claims its new AI system manages to outperform the European model on forecasts out to at least a week and often beyond. DeepMind's system, called GenCast, merges some computational approaches used by atmospheric scientists with a diffusion model, commonly used in generative AI. The result is a system that maintains high resolution while cutting the computational cost significantly. Traditional computational methods have two main advantages over AI systems. The first is that they're directly based on atmospheric physics, incorporating the rules we know govern the behavior of our actual weather, and they calculate some of the details in a way that's directly informed by empirical data. They're also run as ensembles, meaning that multiple instances of the model are run. Due to the chaotic nature of the weather, these different runs will gradually diverge, providing a measure of the uncertainty of the forecast. At least one attempt has been made to merge some of the aspects of traditional weather models with AI systems. An internal Google project used a traditional atmospheric circulation model that divided the Earth's surface into a grid of cells but used an AI to predict the behavior of each cell. This provided much better computational performance, but at the expense of relatively large grid cells, which resulted in relatively low resolution. For its take on AI weather predictions, DeepMind decided to skip the physics and instead adopt the ability to run an ensemble. Gen Cast is based on diffusion models, which have a key feature that's useful here. In essence, these models are trained by starting them with a mixture of an original -- image, text, weather pattern -- and then a variation where noise is injected. The system is supposed to create a variation of the noisy version that is closer to the original. Once trained, it can be fed pure noise and evolve the noise to be closer to whatever it's targeting. In this case, the target is realistic weather data, and the system takes an input of pure noise and evolves it based on the atmosphere's current state and its recent history. For longer-range forecasts, the "history" includes both the actual data and the predicted data from earlier forecasts. The system moves forward in 12-hour steps, so the forecast for day three will incorporate the starting conditions, the earlier history, and the two forecasts from days one and two. This is useful for creating an ensemble forecast because you can feed it different patterns of noise as input, and each will produce a slightly different output of weather data. This serves the same purpose it does in a traditional weather model: providing a measure of the uncertainty for the forecast. For each grid square, GenCast works with six weather measures at the surface, along with six that track the state of the atmosphere and 13 different altitudes at which it estimates the air pressure. Each of these grid squares is 0.2 degrees on a side, a higher resolution than the European model uses for its forecasts. Despite that resolution, DeepMind estimates that a single instance (meaning not a full ensemble) can be run out to 15 days on one of Google's tensor processing systems in just eight minutes. It's possible to make an ensemble forecast by running multiple versions of this in parallel and then integrating the results. Given the amount of hardware Google has at its disposal, the whole process from start to finish is likely to take less than 20 minutes. The source and training data will be placed on the Github page for DeepMind's GraphCast project. Given the relatively low computational requirements, we can probably expect individual academic research teams to start experimenting with it. DeepMind reports that GenCast dramatically outperforms the best traditional forecasting model. Using a standard benchmark in the field, DeepMind found that GenCast was more accurate than the European model on 97 percent of the tests it used, which checked different output values at different times in the future. In addition, the confidence values, based on the uncertainty obtained from the ensemble, were generally reasonable. Past AI weather forecasters, having been trained on real-world data, are generally not great at handling extreme weather since it shows up so rarely in the training set. But GenCast did quite well, often outperforming the European model in things like abnormally high and low temperatures and air pressure (one percent frequency or less, including at the 0.01 percentile). DeepMind also went beyond standard tests to determine whether GenCast might be useful. This research included projecting the tracks of tropical cyclones, an important job for forecasting models. For the first four days, GenCast was significantly more accurate than the European model, and it maintained its lead out to about a week. One of DeepMind's most interesting tests was checking the global forecast of wind power output based on information from the Global Powerplant Database. This involved using it to forecast wind speeds at 10 meters above the surface (which is actually lower than where most turbines reside but is the best approximation possible), then using that number to figure out how much power would be generated. The system beat the traditional weather model by 20 percent for the first two days and stayed in front with a declining lead out to a week. The researchers don't spend much time examining why performance seems to decline gradually out to about a week. Ideally, more details about GenCast's limitations would help inform further improvements, so the researchers are likely thinking about it. In any case, today's paper marks the second case where taking something akin to a hybrid approach -- mixing aspects of traditional forecast systems with AI -- has been reported to improve forecasts. And both those cases took very different approaches, raising the prospect that it will be possible to combine some of their features.

Google's GenCast Might Have Outperformed Top Weather Prediction Systems

Google DeepMind's GenCast is a diffusion model GenCast is trained on weather data from ECMWF's ERA5 archive The AI model can run on the Google Cloud TPU v5 Google introduced its weather predicting artificial intelligence (AI) model GenCast on Wednesday. The AI model was developed by the Mountain View-based tech giant's AI research division Google DeepMind. The company's researchers have also published a paper on the technology highlighting its capabilities in making medium-range weather forecasts. The company claims that the system was able to outperform existing state-of-the-art forecasting models in terms of resolution and accuracy. Notably, GenCast can make weather predictions for the next 15 days with a resolution of 0.25 degrees Celsius. In a blog post, Google DeepMind detailed the new high resolution AI ensemble model. Highlighting that GenCast can make predictions for day-to-day weather and extreme events, it claimed that it was able to perform better than the European Centre for Medium-Range Weather Forecasts' (ECMWF) Ensemble (ENS) system. The performance of the model is now published in the Nature journal. Notably, instead of using the traditional deterministic approach to predict weather, GenCast uses the probabilistic approach. Weather prediction models based on the deterministic approach produce a single, specific forecast for a given set of initial conditions and rely on precise equations of atmospheric physics and chemistry. On the other hand, models based on probabilistic approach generate multiple possible outcomes by simulating a range of initial conditions and model parameters. This is also called ensemble forecasting. Google DeepMind highlighted that GenCast is a diffusion model that adapts to the spherical geometry of the Earth, and generates the complex probability distribution of future weather scenarios. To train the AI model, researchers provided four decades of historical weather data from ECMWF's ERA5 archive. With this, the model was taught global weather patterns at 0.25 degree Celsius resolution. In the published research paper, Google evaluated GenCast's performance by training it on the historical data up to 2018 and then asked it to make predictions for 2019. A total of 1320 combinations across different variables at different lead times were used and the researchers found that GenCast was more accurate than ENS on 97.2 percent of these targets, and on 99.8 percent at lead times greater than 36 hours. Notably, Google DeepMind announced that it will be releasing GenCast AI model's code, weights, and forecasts to support the weather forecasting community.

Google's GenCast AI weather model is better than today's top forecast system

Google has been working on AI weather models for the past few years, and GenCast is its latest forecasting system with "state-of-the-art accuracy." GenCast is a diffusion model -- like what powers image, video, and music generation tools -- but "adapted to the spherical geometry of the Earth, and learns to accurately generate the complex probability distribution of future weather scenarios when given the most recent state of the weather as input," which will continue to be supplied by today's traditional approaches. Google's previous model (GraphCast from 2023) "provided a single, best estimate of future weather." In comparison, GenCast "comprises an ensemble of 50 or more predictions, each representing a possible weather trajectory." An ensemble forecast expresses uncertainty by making multiple predictions that represent different possible scenarios. If most predictions show a cyclone hitting the same area, uncertainty is low. But if they predict different locations, uncertainty is higher. GenCast strikes the right balance, avoiding both overstating or understating its confidence in its forecasts. In practice, Google says GenCast "provides better forecasts of both day-to-day weather and extreme events than the top operational system, the European Centre for Medium-Range Weather Forecasts' (ECMWF) ENS, up to 15 days in advance." Google trained GenCast on historical weather up to 2018 and then tested it on 2019 data. Of 1320 combinations, "GenCast was more accurate than ENS on 97.2% of these targets, and on 99.8% at lead times greater than 36 hours." It's also better at predicting extreme heat and cold, as well as high wind speeds: Now consider tropical cyclones, also known as hurricanes and typhoons. Getting better and more advanced warnings of where they'll strike land is invaluable. GenCast delivers superior predictions of the tracks of these deadly storms. Notably, it takes 8 minutes on a single Google Cloud TPU v5 to create a 15-day forecast with all the possibilities. Existing systems take "hours on a supercomputer with tens of thousands of processors." Google has made GenCast an open model with code and weights available today.

Google says AI weather model masters 15-day forecast

A new artificial intelligence-based weather model can deliver 15-day forecasts with unrivaled accuracy and speed, a Google lab said, with potentially life-saving applications as climate change ramps up. GenCast, invented by London-based AI research laboratory Google DeepMind, "showed better forecasting skill" than the current world-leading model, the company said Wednesday. The European Center for Medium-Range Weather Forecasts (ECMWF) produces predictions for 35 countries and is considered the global benchmark for meteorological accuracy. But DeepMind said GenCast surpassed the precision of the center's forecasts in more than 97 percent of the 1,320 real-world scenarios from 2019 which they were both tested on. Details of its findings were published in Nature, a leading science journal. ECMWF chief Florence Rabier told AFP the project was a "first step" towards integrating AI in weather forecasting but that "it is indeed a leap forward." At this stage it can be used to supplement their current models, she said. "We are progressing year by year," she added. "Any new method that can enhance and accelerate this progress is extremely welcome in the context of the extreme societal pressures of climate change." Extreme weather The model was trained on four decades of temperature, wind speed and air pressure data from 1979 to 2018 and can produce a 15-day forecast in just eight minutes -- compared to the hours it currently takes. "GenCast provides better forecasts of both day-to-day weather and extreme events than the top operational system... up to 15 days in advance," a DeepMind statement said. DeepMind said GenCast "consistently outperformed" the current leading forecast model when predicting extreme heat, extreme cold and high wind speeds. "More accurate forecasts of risks of extreme weather can help officials safeguard more lives, avert damage, and save money," DeepMind said. Extreme weather is becoming more common and more severe as a result of human caused climate change. In August 2023, a series of wildfires in Hawaii killed around 100 people. Authorities were criticized by locals who said they were given no warning of the impending blaze. This summer, a sudden heat wave in Morocco killed at least 21 people over a 24-hour period. And in September, Hurricane Helene killed 237 people in Florida and other southeastern US states. "I am confident that AI-based weather forecasting systems will continue to incrementally improve in the future, including better prediction of extreme events and their intensity, for which there is a lot of need to improve upon," said David Schultz, a professor of synoptic meteorology at Manchester University, who was not involved in the research. But he said these forecasting systems are reliant on the weather prediction models that are already running, such as that operated by ECMWF.

Google says its AI agent outperformed the world's best weather predictions

Google said Wednesday that its own artificial intelligence (AI) agent outperformed the world's best weather predictions. In a blog post, Ilan Price and Matthew Willson, researchers with Google's DeepMind, said its recently-made "AI ensemble model" named GenCast "provides better forecasts of both day-to-day weather and extreme events than the top operational system, the European Centre for Medium-Range Weather Forecasts' (ECMWF) ENS, up to 15 days in advance." Wilson and Price said in their post that they taught GenCast "on historical weather data up to 2018" when trying to analyze the skills of the model "and tested it on data from 2019." "GenCast showed better forecasting skill than ECMWF's ENS, the top operational ensemble forecasting system that many national and local decisions depend upon every day," the researchers said. The researchers said they assessed the capabilities of ECMWF's ENS and GenCast through examining "forecasts of different variables at different lead times -- 1320 combinations in total." According to the American Meteorological Society, a forecast lead time is "the length of time between the issuance of a forecast and the occurrence of the phenomena that were predicted." Among the variables tested were wind speed and temperature, the blog post read. According to the DeepMind researchers, their system beat out ENS on accuracy 97.2 percent of the time when it came to the forecasts of different variables at different lead times. When lead times were over 36 hours, they said, GenCast beat out ENS for accuracy on forecasts of different variables at the different lead times 99.8 percent of the time. Wilson and Price said that despite the success of GenCast, "traditional models remain essential for" forecasting because "they supply the training data and initial weather conditions required by models such as GenCast." "This cooperation between AI and traditional meteorology highlights the power of a combined approach to improve forecasts and better serve society," the researchers said. The Hill has reached out to ECMWF for further comment.

Google says its new AI model outperforms the top weather forecast system

Google's DeepMind team unveiled an AI model for weather prediction this week called GenCast. In a paper published in Nature, DeepMind researchers said they found that GenCast outperforms the European Centre for Medium-Range Weather Forecasts' ENS -- apparently the world's top operational forecasting system. And in a blog post, the DeepMind team offered a more accessible explanation of the tech: While its previous weather model was "deterministic, and provided a single, best estimate of future weather," GenCast "comprises an ensemble of 50 or more predictions, each representing a possible weather trajectory," creating a "complex probability distribution of future weather scenarios." As for how it stacks up against ENS, the team said it trained GenCast on weather data up to 2018, then compared its forecasts for 2019, finding that GenCast was more accurate 97.2 percent of the time. Google says GenCast is part of its suite of AI-based weather models, which it's starting to incorporate into Google Search and Maps. It also plans to release real-time and historical forecasts from GenCast, which anyone can use into their own research and models.

Here's How Google's New AI Weather Tool Could Help Business Forecasts, Too

In many instances, it did a more accurate job at predicting the weather than one of the existing, leading computer models, one often cited as the most accurate option. Given that many businesses daily operations and strategic plans are directly impacted by weather conditions, and severe weather events can affect huge numbers of companies, super-accurate, AI-powered predictions could save many of them a lot of money in the future. GenCast was tested against the ENS weather prediction model run by the European Centre for Medium-Range Weather Forecasts, news site The Verge reported. ENS is, the Verge says "one of the world's top-tier models for forecasting." When it was tested in real world conditions, GenCast outperformed ENS 97.2 percent of the time. Like many generative AI models, GenCast works by being trained on reams upon reams of real data, which are used to shape the algorithms that make the AI tick. In this case that database was built from data gathered from 1979 to 2018 -- this information helped GenCast understand the subtle interactions between different parts of the environment and climate that cause, say, rain to hit one location under certain conditions, or snow to arrive under slightly different ones. As the Verge points out, this is really different from how current weather models work: these rely on carefully hand-tuned mathematical models that run on supercomputers, basically crunching the numbers on what atmospheric physics and ground-based effects are going to cause which type of weather.

Google DeepMind unveils GenCast AI model for weather forecasting

Google DeepMind on Wednesday introduced GenCast, a high-resolution AI ensemble model designed to enhance weather forecasting accuracy. Capable of predicting daily and extreme weather events up to 15 days in advance, GenCast outperforms the European Centre for Medium-Range Weather Forecasts' (ECMWF) ENS system. AI-Powered Ensemble Forecasting GenCast generates forecasts using an ensemble of 50+ predictions, offering multiple possible weather scenarios. This approach accounts for uncertainties, providing a range of outcomes instead of a single prediction. Diffusion Model Design GenCast is based on a diffusion model, similar to those used in image and video generation. However, it is uniquely tailored to Earth's spherical geometry, enabling accurate mapping of complex weather patterns based on current data. Training with Decades of Historical Data The model was trained with 40 years of data from ECMWF's ERA5 archive, including variables such as temperature, wind speed, and atmospheric pressure across various altitudes. High-Resolution Accuracy Operating at a resolution of 0.25 degrees, GenCast delivers precise predictions of weather patterns and extreme events. It outperformed ENS in 97.2% of cases and achieved 99.8% accuracy for forecasts beyond 36 hours. Using a single Google Cloud TPU v5, GenCast generates a 15-day forecast in just 8 minutes. Forecasts run in parallel, drastically reducing computation time compared to traditional methods that require supercomputers and hours of processing. GenCast is part of Google's broader AI initiative, which includes tools like deterministic medium-range forecasts, NeuralGCM, SEEDS, and flooding models. These models are already integrated into Google Search and Maps, enhancing predictions for precipitation, wildfires, extreme heat, and flooding. Google emphasized the importance of working with weather agencies, meteorologists, and researchers. They plan to continue developing AI-based forecasting methods while retaining traditional models to provide critical training data and initial conditions. These collaborations are essential for improving global weather predictions. To foster innovation, Google has made GenCast an open model, releasing its code and weights to the public. They also plan to release real-time and historical forecasts, enabling researchers and organizations to incorporate GenCast into their models. Google expressed eagerness to engage with the global weather community, emphasizing partnerships that can accelerate research and enhance the model's impact across commercial and non-commercial sectors.

AI weather models can now beat the best traditional forecasts

CSIRO provides funding as a founding partner of The Conversation AU. A new machine-learning weather prediction model called GenCast can outperform the best traditional forecasting systems in at least some situations, according to a paper by Google DeepMind researchers published today in Nature. Using a diffusion model approach similar to artificial intelligence (AI) image generators, the system generates multiple forecasts to capture the complex behaviour of the atmosphere. It does so with a fraction of the time and computing resources required for traditional approaches. How weather forecasts work The weather predictions we use in practice are produced by running multiple numerical simulations of the atmosphere. Each simulation starts from a slightly different estimate of the current weather. This is because we don't know exactly what the weather is at this instant everywhere in the world. To know that, we would need sensor measurements everywhere. These numerical simulations use a model of the world's atmosphere divided into a grid of three-dimensional blocks. By solving equations describing the fundamental physical laws of nature, the simulations predict what will happen in the atmosphere. Known as general circulation models, these simulations need a lot of computing power. They are usually run at high-performance supercomputing facilities. Machine-learning the weather The past few years have seen an explosion in efforts to produce weather prediction models using machine learning. Typically, these approaches don't incorporate our knowledge of the laws of nature the way general circulation models do. Most of these models use some form of neural network to learn patterns in historical data and produce a single future forecast. However, this approach produces predictions that lose detail as they progress into the future, gradually becoming "smoother". This smoothness is not what we see in real weather systems. Researchers at Google's DeepMind AI research lab have just published a paper in Nature describing their latest machine-learning model, GenCast. GenCast mitigates this smoothing effect by generating an ensemble of multiple forecasts. Each individual forecast is less smooth, and better resembles the complexity observed in nature. The best estimate of the actual future then comes from averaging the different forecasts. The size of the differences between the individual forecasts indicates how much uncertainty there is. According to the GenCast paper, this probabilistic approach creates more accurate forecasts than the best numerical weather prediction system in the world - the one at the European Centre for Medium-Range Weather Forecasts. Generative AI - for weather GenCast is trained on what is called reanalysis data from the years 1979 to 2018. This data is produced by the kind of general circulation models we talked about earlier, which are additionally corrected to resemble actual historical weather observations to produce a more consistent picture of the world's weather. The GenCast model makes predictions of several variables such as temperature, pressure, humidity and wind speed at the surface and at 13 different heights, on a grid that divides the world up into 0.25-degree regions of latitude and longitude. GenCast is what is called a "diffusion model", similar to AI image generators. However, instead of taking text and producing an image, it takes the current state of the atmosphere and produces an estimate of what it will be like in 12 hours. This works by first setting the values of the atmospheric variables 12 hours into the future as random noise. GenCast then uses a neural network to find structures in the noise that are compatible with the current and previous weather variables. An ensemble of multiple forecasts can be generated by starting with different random noise. Forecasts are run out to 15 days, taking 8 minutes on a single processor called a tensor processor unit (TPU). This is significantly faster than a general circulation model. The training of the model took five days using 32 TPUs. Machine-learning forecasts could become more widespread in the coming years as they become more efficient and reliable. However, classical numerical weather prediction and reanalysed data will still be required. Not only are they needed to provide the initial conditions for the machine learning weather forecasts, they also produce the input data to continually fine-tune the machine learning models. What about the climate? Current machine learning weather forecasting systems are not appropriate for climate projections, for three reasons. Firstly, to make weather predictions weeks into the future, you can assume that the ocean, land and sea ice won't change. This is not the case for climate predictions over multiple decades. Secondly, weather prediction is highly dependent on the details of the current weather. However, climate projections are concerned with the statistics of the climate decades into the future, for which today's weather is irrelevant. Future carbon emissions are the greater determinant of the future state of the climate. Thirdly, weather prediction is a "big data" problem. There are vast amounts of relevant observational data, which is what you need to train a complex machine learning model. Climate projection is a "small data" problem, with relatively little available data. This is because the relevant physical phenomena (such as sea levels or climate drivers such as the El Niño-Southern Oscillation) evolve much more slowly than the weather. There are ways to address these problems. One approach is to use our knowledge of physics to simplify our models, meaning they require less data for machine learning. Another approach is to use physics-informed neural networks to try to fit the data and also satisfy the laws of nature. A third is to use physics to set "ground rules" for a system, then use machine learning to determine the specific model parameters. Machine learning has a role to play in the future of both weather forecasting and climate projections. However, fundamental physics - fluid mechanics and thermodynamics - will continue to play a crucial role.

GenCast predicts weather and the risks of extreme conditions with state-of-the-art accuracy

New AI model advances the prediction of weather uncertainties and risks, delivering faster, more accurate forecasts up to 15 days ahead Weather impacts all of us -- shaping our decisions, our safety, and our way of life. As climate change drives more extreme weather events, accurate and trustworthy forecasts are more essential than ever. Yet, weather cannot be predicted perfectly, and forecasts are especially uncertain beyond a few days. Because a perfect weather forecast is not possible, scientists and weather agencies use probabilistic ensemble forecasts, where the model predicts a range of likely weather scenarios. Such ensemble forecasts are more useful than relying on a single forecast, as they provide decision makers with a fuller picture of possible weather conditions in the coming days and weeks and how likely each scenario is. Today, in a paper published in Nature, we present GenCast, our new high resolution (0.25°) AI ensemble model. GenCast provides better forecasts of both day-to-day weather and extreme events than the top operational system, the European Centre for Medium-Range Weather Forecasts' (ECMWF) ENS, up to 15 days in advance. We'll be releasing our model's code, weights, and forecasts, to support the wider weather forecasting community.

Google introduces AI agent that aces 15-day weather forecasts

Now, a new artificial intelligence tool from DeepMind, a Google company in London that develops AI applications, has smashed through the old barriers and achieved what its makers call unmatched skill and speed in devising 15-day weather forecasts. They report in the journal Nature on Wednesday that their new model can, among other things, outperform the world's best forecasts meant to track deadly storms and save lives.In the 1960s, weather scientists found that the chaotic nature of Earth's atmosphere would put a limit on how far into the future their forecasts might peer. Two weeks seemed to be the limit. Still, by the early 2000s, the great difficulty of the undertaking kept reliable forecasts restricted to about a week. Now, a new artificial intelligence tool from DeepMind, a Google company in London that develops AI applications, has smashed through the old barriers and achieved what its makers call unmatched skill and speed in devising 15-day weather forecasts. They report in the journal Nature on Wednesday that their new model can, among other things, outperform the world's best forecasts meant to track deadly storms and save lives. "It's a big deal," said Kerry Emanuel, a professor emeritus of atmospheric science at the Massachusetts Institute of Technology who was not involved in the DeepMind research. "It's an important step forward." In 2019, Emanuel and six other experts, writing in the Journal of the Atmospheric Sciences, argued that advancing the development of reliable forecasts to a length of 15 days from 10 days would have "enormous socioeconomic benefits" by helping the public avoid the worst effects of extreme weather. Ilan Price, the new paper's lead author and a senior research scientist at DeepMind, described the new AI agent, which the team calls GenCast, as much faster than traditional methods. "And it's more accurate," he added. He and his colleagues found that GenCast ran circles around DeepMind's previous AI weather program, which debuted in late 2023 with reliable 10-day forecasts. Rémi Lam, the lead scientist on that project and one of a dozen co-authors on the new paper, described the company's weather team as having made surprisingly fast progress. "I'm a little bit reluctant to say it, but it's like we've made decades worth of improvements in one year," he said in an interview. "We're seeing really, really rapid progress." The world leader in atmospheric prediction is the European Center for Medium-Range Weather Forecasts. Comparative tests regularly show that its projections exceed all others in accuracy. DeepMind tested its new AI program against the center's Ensemble Prediction System -- a service that 35 nations rely on to produce their own weather forecasts. The team compared how the 15-day forecasts of both systems performed in predicting a designated set of 1,320 global wind speeds, temperatures and other atmospheric features. The Nature report said the new agent outdid the center's forecasts 97.2% of time. The AI achievement, the authors wrote, "helps open the next chapter in operational weather forecasting." Matthew Chantry, an AI specialist at the European Center for Medium-Range Weather Forecasts, said his agency was already adopting some of its features. "That's how highly we think of it," he said. Machine learning in general, Chantry added, was accelerating human bids to outmaneuver some of nature's deadliest threats. DeepMind's weather advance comes two months after other AI researchers in the company shared the Nobel Prize for chemistry. The scientific news forms a bright counterpoint to public fears of AI stealing jobs and driving humans to the edge of obsolescence. The natural chaos in Earth's atmosphere means that all weather forecasts, including the two-week variety, grow less reliable as they peer further into the future. Even so, AccuWeather offers 90-day forecasts. And the Old Farmer's Almanac says it can gaze ahead 60 days. DeepMind backs its 15-day declaration with pages of evidence laid out in one of the world's leading science journals, Nature. So too, Google posted an online blog that details the AI advance. The new GenCast agent takes a radically different approach from mainstream forecasting, which uses room-size supercomputers that turn millions of global observations and calculations into predictions. Instead, the DeepMind agent runs on smaller machines and studies the atmospheric patterns of the past to learn the subtle dynamics that result in the planet's weather. The DeepMind team trained GenCast on a massive archive of weather data curated by the European center. The training period went from 1979 to 2018, or 40 years. The team then tested how well the agent could predict 2019's weather. Such training empowers all types of generative AI -- the kind that's creative. Mimicking how humans learn, it spots patterns in mountains of data and then makes new, original material that has similar characteristics. Lam of DeepMind noted that GenCast's generative skills were rooted in factual data gathered from nature rather than the internet, notorious for its confusing mix of facts, biases and fallacies. "We have a ground truth," he said of its dependence on natural phenomena. "We have a reality check." The new agent's forecasts are probabilistic -- like those on the weather apps of smartphones. For instance, GenCast can give a range of percentages for the likelihood of rain in a specific region on a given day. In contrast, its DeepMind predecessor, GraphCast, offers a single forecast for a particular time and location. Known as deterministic, its method is essentially a best guess that gives no indication of the prediction's uncertainty. Probabilistic forecasts are considered more nuanced and sophisticated than the deterministic kind, and are more difficult to create. Typically, a GenCast forecast draws from a set of 50 or more predictions that produce its range of probabilities. Despite all the effort that goes into those calculations, Price of DeepMind said, the new agent can generate a 15-day forecast in minutes compared with hours for a supercomputer. That can make its projections much timelier -- an advantage in tracking fast-moving storms. GenCast, the team says, can predict with great accuracy the paths of hurricanes, which annually can take thousands of lives and rack up hundreds of billions of dollars in property damage. The Nature paper said comparative testing showed that its hurricane track predictions consistently outdid those of the European center. Emanuel of MIT said the DeepMind team failed to mention that its new agent provides little information about hurricane intensity. Price, the paper's lead author, concurred. He said the problem lay in training data limitations on hurricane wind speed. The weather team, he added, was confident it could devise a solution. GenCast will most likely complement current methods rather than replace them, Emanuel argued. Each type, he said, has its own strengths and weaknesses in predicting the riot of variable phenomena that constitute the weather. "The status quo isn't going to disappear," Emanuel said. "Perhaps the two of them working together will prove to be the best way forward." For its part, the DeepMind team acknowledged its heavy reliance on the conventional world of weather readings -- noting, for instance, how its AI training data comes from the giant European weather archive. Its computations also start with a snapshot of the world's current weather, what the team calls initial conditions. The team hopes that other weather experts will test its new technology. Price said that the DeepMind team would share online its AI agent and underlying computer code. He added that GenCast's weather predictions would soon be posted publicly on Google's Earth Engine and Big Query, giving scientists access to the new forecasts. "We're excited for the community to use and build on our research," Price said. Chantry of the European center said Google and DeepMind might have hidden their AI advance behind a wall of corporate secrecy, using it "to make a better weather forecast for their own apps and telling no one how they did it." Instead, he added, the emerging field has embraced a public openness that's helping "lots and lots of people engage in this revolution."

Google says AI weather model masters 15-day forecast

San Francisco (AFP) - A new artificial intelligence-based weather model can deliver 15-day forecasts with unrivaled accuracy and speed, a Google lab said, with potentially life-saving applications as climate change ramps up. GenCast, invented by London-based AI research laboratory Google DeepMind, "showed better forecasting skill" than the current world-leading model, the company said Wednesday. The European Centre for Medium-Range Weather Forecasts (ECMWF) produces predictions for 35 countries and is considered the global benchmark for meteorological accuracy. But DeepMind said GenCast surpassed the precision of the center's forecasts in more than 97 percent of the 1,320 real-world scenarios from 2019 which they were both tested on. Details of its findings were published in Nature, a leading science journal. ECMWF chief Florence Rabier told AFP the project was a "first step" towards integrating AI in weather forecasting but that "it is indeed a leap forward." At this stage it can be used to supplement their current models, she said. "We are progressing year by year," she added. "Any new method that can enhance and accelerate this progress is extremely welcome in the context of the extreme societal pressures of climate change." The model was trained on four decades of temperature, wind speed and air pressure data from 1979 to 2018 and can produce a 15-day forecast in just eight minutes -- compared to the hours it currently takes. "GenCast provides better forecasts of both day-to-day weather and extreme events than the top operational system... up to 15 days in advance," a DeepMind statement said. Scientists have warned extreme weather is becoming more common and more severe as a result of man made climate change. Last August, a series of wildfires in Hawaii killed around 100 people. Authorities were criticized by locals who said they were given no warning of the impending blaze. This summer, a sudden heatwave in Morocco killed at least 21 people over a 24-hour period. And in September, Hurricane Helene killed 237 people in Florida and other southeastern US states. DeepMind said GenCast "consistently outperformed" the current leading forecast model when predicting extreme heat, extreme cold and high wind speeds. "More accurate forecasts of risks of extreme weather can help officials safeguard more lives, avert damage, and save money," DeepMind said.

GenCast: Our new AI model provides more accurate weather results, faster.

In a paper published in Nature, Google DeepMind introduced its newest AI model, GenCast, which can predict weather and the risks of extreme conditions with state-of-the-art accuracy, delivering faster and more accurate forecasts up to 15 days ahead. More accurate weather predictions can help officials prepare for natural disasters, which can save lives and protect homes and communities. To read more about GenCast and other weather prediction models from Google, check out the Google DeepMind blog.

Google introduces AI agent that aces 15-day weather forecasts

In the 1960s, weather scientists found that the chaotic nature of Earth's atmosphere would put a limit on how far into the future their forecasts might peer. Two weeks seemed to be the limit. Still, by the early 2000s, the great difficulty of the undertaking kept reliable forecasts restricted to about a week. Now, a new artificial intelligence tool from DeepMind, a Google company in London that develops AI applications, has smashed through the old barriers and achieved what its makers call unmatched skill and speed in devising 15-day weather forecasts. They report in the journal Nature on Wednesday that their new model can, among other things, outperform the world's best forecasts meant to track deadly storms and save lives. "It's a big deal," said Kerry Emanuel, a professor emeritus of atmospheric science at the Massachusetts Institute of Technology who was not involved in the DeepMind research. "It's an important step forward." In 2019, Emanuel and six other experts, writing in the Journal of the Atmospheric Sciences, argued that advancing the development of reliable forecasts to a length of 15 days from 10 days would have "enormous socioeconomic benefits" by helping the public avoid the worst effects of extreme weather. Ilan Price, the new paper's lead author and a senior research scientist at DeepMind, described the new AI agent, which the team calls GenCast, as much faster than traditional methods. "And it's more accurate," he added. He and his colleagues found that GenCast ran circles around DeepMind's previous AI weather program, which debuted in late 2023 with reliable 10-day forecasts. Rémi Lam, the lead scientist on that project and one of a dozen co-authors on the new paper, described the company's weather team as having made surprisingly fast progress. "I'm a little bit reluctant to say it, but it's like we've made decades worth of improvements in one year," he said in an interview. "We're seeing really, really rapid progress." The world leader in atmospheric prediction is the European Center for Medium-Range Weather Forecasts. Comparative tests regularly show that its projections exceed all others in accuracy. DeepMind tested its new AI program against the center's Ensemble Prediction System -- a service that 35 nations rely on to produce their own weather forecasts. The team compared how the 15-day forecasts of both systems performed in predicting a designated set of 1,320 global wind speeds, temperatures and other atmospheric features. The Nature report said the new agent outdid the center's forecasts 97.2% of time. The AI achievement, the authors wrote, "helps open the next chapter in operational weather forecasting." Matthew Chantry, an AI specialist at the European Center for Medium-Range Weather Forecasts, said his agency was already adopting some of its features. "That's how highly we think of it," he said. Machine learning in general, Chantry added, was accelerating human bids to outmaneuver some of nature's deadliest threats. DeepMind's weather advance comes two months after other AI researchers in the company shared the Nobel Prize for chemistry. The scientific news forms a bright counterpoint to public fears of AI stealing jobs and driving humans to the edge of obsolescence. The natural chaos in Earth's atmosphere means that all weather forecasts, including the two-week variety, grow less reliable as they peer further into the future. Even so, AccuWeather offers 90-day forecasts. And the Old Farmer's Almanac says it can gaze ahead 60 days. DeepMind backs its 15-day declaration with pages of evidence laid out in one of the world's leading science journals, Nature. So too, Google posted an online blog that details the AI advance. The new GenCast agent takes a radically different approach from mainstream forecasting, which uses room-size supercomputers that turn millions of global observations and calculations into predictions. Instead, the DeepMind agent runs on smaller machines and studies the atmospheric patterns of the past to learn the subtle dynamics that result in the planet's weather. The DeepMind team trained GenCast on a massive archive of weather data curated by the European center. The training period went from 1979 to 2018, or 40 years. The team then tested how well the agent could predict 2019's weather. Such training empowers all types of generative AI -- the kind that's creative. Mimicking how humans learn, it spots patterns in mountains of data and then makes new, original material that has similar characteristics. Lam of DeepMind noted that GenCast's generative skills were rooted in factual data gathered from nature rather than the internet, notorious for its confusing mix of facts, biases and fallacies. "We have a ground truth," he said of its dependence on natural phenomena. "We have a reality check." The new agent's forecasts are probabilistic -- like those on the weather apps of smartphones. For instance, GenCast can give a range of percentages for the likelihood of rain in a specific region on a given day. In contrast, its DeepMind predecessor, GraphCast, offers a single forecast for a particular time and location. Known as deterministic, its method is essentially a best guess that gives no indication of the prediction's uncertainty. Probabilistic forecasts are considered more nuanced and sophisticated than the deterministic kind, and are more difficult to create. Typically, a GenCast forecast draws from a set of 50 or more predictions that produce its range of probabilities. Despite all the effort that goes into those calculations, Price of DeepMind said, the new agent can generate a 15-day forecast in minutes compared with hours for a supercomputer. That can make its projections much timelier -- an advantage in tracking fast-moving storms. GenCast, the team says, can predict with great accuracy the paths of hurricanes, which annually can take thousands of lives and rack up hundreds of billions of dollars in property damage. The Nature paper said comparative testing showed that its hurricane track predictions consistently outdid those of the European center. Emanuel of MIT said the DeepMind team failed to mention that its new agent provides little information about hurricane intensity. Price, the paper's lead author, concurred. He said the problem lay in training data limitations on hurricane wind speed. The weather team, he added, was confident it could devise a solution. GenCast will most likely complement current methods rather than replace them, Emanuel argued. Each type, he said, has its own strengths and weaknesses in predicting the riot of variable phenomena that constitute the weather. "The status quo isn't going to disappear," Emanuel said. "Perhaps the two of them working together will prove to be the best way forward." For its part, the DeepMind team acknowledged its heavy reliance on the conventional world of weather readings -- noting, for instance, how its AI training data comes from the giant European weather archive. Its computations also start with a snapshot of the world's current weather, what the team calls initial conditions. The team hopes that other weather experts will test its new technology. Price said that the DeepMind team would share online its AI agent and underlying computer code. He added that GenCast's weather predictions would soon be posted publicly on Google's Earth Engine and Big Query, giving scientists access to the new forecasts. "We're excited for the community to use and build on our research," Price said. Chantry of the European center said Google and DeepMind might have hidden their AI advance behind a wall of corporate secrecy, using it "to make a better weather forecast for their own apps and telling no one how they did it." Instead, he added, the emerging field has embraced a public openness that's helping "lots and lots of people engage in this revolution."

Google Introduces A.I. Agent That Aces 15-Day Weather Forecasts

William J. Broad reported earlier this year on another artificial intelligence weather forecasting agent from Google DeepMind. In the 1960s, weather scientists found that the chaotic nature of Earth's atmosphere would put a limit on how far into the future their forecasts might peer. Two weeks seemed to be the limit. Still, by the early 2000s, the great difficulty of the undertaking kept reliable forecasts restricted to about a week. Now, a new artificial intelligence tool from DeepMind, a Google company in London that develops A.I. applications, has smashed through the old barriers and achieved what its makers call unmatched skill and speed in devising 15-day weather forecasts. They report in the journal Nature on Wednesday that their new model can, among other things, outperform the world's best forecasts meant to track deadly storms and save lives. "It's a big deal," said Kerry Emanuel, a professor emeritus of atmospheric science at the Massachusetts Institute of Technology who was not involved in the DeepMind research. "It's an important step forward." In 2019, Dr. Emanuel and six other experts, writing in the Journal of the Atmospheric Sciences, argued that advancing the development of reliable forecasts to a length of 15 days from 10 days would have "enormous socioeconomic benefits" by helping the public avoid the worst effects of extreme weather. Ilan Price, the new paper's lead author and a senior research scientist at DeepMind, described the new A.I. agent, which the team calls GenCast, as much faster than traditional methods. "And it's more accurate," he added. He and his colleagues found that GenCast ran circles around DeepMind's previous A.I. weather program, which debuted in late 2023 with reliable 10-day forecasts. Rémi Lam, the lead scientist on that project and one of a dozen co-authors on the new paper, described the company's weather team as having made surprisingly fast progress. "I'm a little bit reluctant to say it, but it's like we've made decades worth of improvements in one year," he said in an interview. "We're seeing really, really rapid progress." The world leader in atmospheric prediction is the European Center for Medium-Range Weather Forecasts. Comparative tests regularly show that its projections exceed all others in accuracy. DeepMind tested its new A.I. program against the center's Ensemble Prediction System -- a service that 35 nations rely on to produce their own weather forecasts. The team compared how the 15-day forecasts of both systems performed in predicting a designated set of 1,320 global wind speeds, temperatures and other atmospheric features. The Nature report said the new agent outdid the center's forecasts 97.2 percent of time. The A.I. achievement, the authors wrote, "helps open the next chapter in operational weather forecasting." Matthew Chantry, an A.I. specialist at the European Center for Medium-Range Weather Forecasts, said his agency was already adopting some of its features. "That's how highly we think of it," he said. Machine learning in general, Dr. Chantry added, was accelerating human bids to outmaneuver some of nature's deadliest threats. DeepMind's weather advance comes two months after other A.I. researchers in the company shared the Nobel Prize for chemistry. The scientific news forms a bright counterpoint to public fears of A.I. stealing jobs and driving humans to the edge of obsolescence. The natural chaos in Earth's atmosphere means that all weather forecasts, including the two-week variety, grow less reliable as they peer further into the future. Even so, AccuWeather offers 90-day forecasts. And the Old Farmer's Almanac says it can gaze ahead 60 days. DeepMind backs its 15-day declaration with pages of evidence laid out in one of the world's leading science journals, Nature. So too, Google posted an online blog that details the A.I. advance. The new GenCast agent takes a radically different approach from mainstream forecasting, which uses room-size supercomputers that turn millions of global observations and calculations into predictions. Instead, the DeepMind agent runs on smaller machines and studies the atmospheric patterns of the past to learn the subtle dynamics that result in the planet's weather. The DeepMind team trained GenCast on a massive archive of weather data curated by the European center. The training period went from 1979 to 2018, or 40 years. The team then tested how well the agent could predict 2019's weather. Such training empowers all types of generative A.I. -- the kind that's creative. Mimicking how humans learn, it spots patterns in mountains of data and then makes new, original material that has similar characteristics. Dr. Lam of DeepMind noted that GenCast's generative skills were rooted in factual data gathered from nature rather than the internet, notorious for its confusing mix of facts, biases and fallacies. "We have a ground truth," he said of its dependence on natural phenomena. "We have a reality check." The new agent's forecasts are probabilistic -- like those on the weather apps of smartphones. For instance, GenCast can give a range of percentages for the likelihood of rain in a specific region on a given day. In contrast, its DeepMind predecessor, GraphCast, offers a single forecast for a particular time and location. Known as deterministic, its method is essentially a best guess that gives no indication of the prediction's uncertainty. Probabilistic forecasts are considered more nuanced and sophisticated than the deterministic kind, and are more difficult to create. Typically, a GenCast forecast draws from a set of 50 or more predictions that produce its range of probabilities. Despite all the effort that goes into those calculations, Dr. Price of DeepMind said, the new agent can generate a 15-day forecast in minutes compared with hours for a supercomputer. That can make its projections much timelier -- an advantage in tracking fast-moving storms. GenCast, the team says, can predict with great accuracy the paths of hurricanes, which annually can take thousands of lives and rack up hundreds of billions of dollars in property damage. The Nature paper said comparative testing showed that its hurricane track predictions consistently outdid those of the European center. Dr. Emanuel of M.I.T. said the DeepMind team failed to mention that its new agent provides little information about hurricane intensity. Dr. Price, the paper's lead author, concurred. He said the problem lay in training data limitations on hurricane wind speed. The weather team, he added, was confident it could devise a solution. GenCast will most likely complement current methods rather than replace them, Dr. Emanuel argued. Each type, he said, has its own strengths and weaknesses in predicting the riot of variable phenomena that constitute the weather. "The status quo isn't going to disappear," Dr. Emanuel said. "Perhaps the two of them working together will prove to be the best way forward." For its part, the DeepMind team acknowledged its heavy reliance on the conventional world of weather readings -- noting, for instance, how its A.I. training data comes from the giant European weather archive. Its computations also start with a snapshot of the world's current weather, what the team calls initial conditions. The team hopes that other weather experts will test its new technology. Dr. Price said that the DeepMind team would share online its A.I. agent and underlying computer code. He added that GenCast's weather predictions would soon be posted publicly on Google's Earth Engine and Big Query, giving scientists access to the new forecasts. "We're excited for the community to use and build on our research," Dr. Price said. Dr. Chantry of the European center said Google and DeepMind might have hidden their A.I. advance behind a wall of corporate secrecy, using it "to make a better weather forecast for their own apps and telling no one how they did it." Instead, he added, the emerging field has embraced a public openness that's helping "lots and lots of people engage in this revolution."

AI weatherman: the DeepMind researcher making faster, more accurate forecasts

Rémi Lam had heard about San Francisco's microclimates, but he didn't realize how idiosyncratic they could be until he moved there this year. "The street I live in can be foggy, and it's sunny two blocks down," he says. Weather forecasts for the city can be wildly incorrect depending on the location. Even state-of-the-art weather forecasts can't predict the city's microclimates and how they will vary. Lam has spent a lot of time thinking about weather and how to forecast it. As a researcher at Google DeepMind, the artificial intelligence (AI) firm based in London, Lam has been pioneering the use of machine learning to improve weather prediction. This field has made rapid advances in the past few years, and Lam and his colleagues have been at the forefront of these efforts. They're not alone. A number of groups are racing to develop AI-aided weather forecasts, including those at Microsoft, Nvidia, Huawei and the European Centre for Medium-Range Weather Forecasts (ECMWF) in Reading, UK. But for much of this year, the leading AI in terms of accuracy was a project called GraphCast, led by Lam (R. Lam et al. Science 382, 1416-1421; 2023). "GraphCast raised the bar up in terms of skill of forecasting," says Matthew Chantry, who leads research on AI-based weather prediction at the ECMWF. Conventional weather forecasts are sophisticated programs that simulate the evolution of Earth's atmosphere on the basis of known physics of how air, heat and water vapour move around the planet. GraphCast is an artificial neural network that is shaped like a grid covering the globe. Lam and his collaborators 'trained' it with data based on real atmospheric measurements, but without giving it any explicit knowledge of physical laws. Still, the AI forecasts were by many measures better than the conventional ones. "I was surprised it outperformed the physics-based forecasts so quickly -- I thought it would take longer," says Lam. And although the training is computationally intensive, the forecasts take less than a minute on an advanced desktop computer -- versus the hours of supercomputer running time for conventional ones. Lam was born in a suburb of Paris in 1988, and trained as an aerospace engineer in France and the United States. He then realized that his understanding of the statistical modelling of fluid mechanics could be helpful to those using AI. DeepMind, with a culture focused on solving scientific problems, turned out to be an ideal fit. "There's just no better place to do machine learning," he says. Maria Molina, an atmospheric scientist who applies AI to weather and climate modelling at the University of Maryland in College Park, gives credit to corporations such as Google for making their weather models available for anyone to download and run on their computers -- at least so far. "At some point, when does that goodwill run out?" It could be worrying if those companies some day came to monopolize the best-available forecasts, she adds, especially when it comes to extreme weather events. "We should never expect the public to pay for access to life-saving information." AI systems are not yet able to generate accurate weather forecasts independently, Lam points out. In particular, GraphCast and other tools still rely on data generated by physics-based models as a starting point. But machine-learning techniques in weather forecasting are improving quickly. This month, a team from DeepMind that included Lam published a model called GenCast (I. Price et al. Nature https://doi.org/10.1038/s41586-024-08252-9; 2024). The team says GenCast takes just 8 minutes to produce a group of 15-day forecasts that are more accurate than conventional ones. Lam hopes for the day when machine learning can help forecasters to produce reliable predictions beyond the current limit of ten days or so, and make much more detailed, local predictions -- perhaps even for San Francisco fog.

Google introduces AI agent that aces 15-day weather forecasts

In the 1960s, weather scientists found that the chaotic nature of Earth's atmosphere would put a limit on how far into the future their forecasts might peer. Two weeks seemed to be the limit. Still, by the early 2000s, the great difficulty of the undertaking kept reliable forecasts restricted to about a week. Now, a new artificial intelligence tool from DeepMind, a Google company in London that develops AI applications, has smashed through the old barriers and achieved what its makers call unmatched skill and speed in devising 15-day weather forecasts. They report in the journal Nature on Wednesday that their new model can, among other things, outperform the world's best forecasts meant to track deadly storms and save lives. "It's a big deal," said Kerry Emanuel, a professor emeritus of atmospheric science at the Massachusetts Institute of Technology who was not involved in the DeepMind research. "It's an important step forward."

Google's AI Weather App Could be Extremely Useful to Photographers

Google has introduced a new type of AI-powered weather app capable of predicting the weather earlier than traditional models, which could be handy for photographers. Anyone planning a photo shoot will know the anxiety of hoping that the weather will play nice and Google's GenCast could be a useful tool because it claims to "deliver faster, more accurate forecasts up to 15 days ahead". According to a paper published in Nature, GenCast outperforms one of the world's best forecast models, ENS by the European Centre for Medium-Range Weather Forecasts (ECMWF). GenCast is an AI diffusion model, the same type of technology that powers AI image generators. It was trained on four decades worth of weather data, from 1979 to 2018, including temperatures, wind speed, and pressure at various altitudes. The model takes that data and learns to recognize patterns so it can make predictions about what is likely to happen in the future. The model then made predictions for the year 2019 and beat ENS's predictions 97.2 percent of the time. That number rose to 99.8 percent against predictions made more than 36 hours before. It also works well on extreme weather events. For example. GenCast gave an additional 12 hours of warning for a tropical cyclone and was better at predicting wind power production. Although GenCast was testing itself against the 2019 version of ENS, which has obviously been upgraded since then, it still "marks a significant milestone in the evolution of weather forecasting," according to the ECMWF machine learning coordinator Matt Chantry, per The Verge. GenCast also excels at the speed of forecasting: it can produce a 15-day forecast in just eight minutes versus ENS which can take several hours to produce the same report. "Computationally, it's orders of magnitude more expensive to run traditional forecasts compared to a model like Gencast," says Ilan Price, a senior research scientist at DeepMind. It remains to be seen whether an AI weather model will be useful in the real world but better longer-term forecasts is something that everyone, not just photographers, would be grateful for.

Google DeepMind's latest AI models can forecast the weather with incredible accuracy and generate playable 3D worlds - SiliconANGLE

Google DeepMind's latest AI models can forecast the weather with incredible accuracy and generate playable 3D worlds Google LLC's artificial intelligence research organization DeepMind has announced the release of two very different, but exceedingly powerful large language models. One of them, GenCast, is focused on weather forecasts and the other, called Genie 2, is designed to create elaborate, visually striking virtual worlds for video games. GenCast appears to have the most practical implications. Announced in a blog post by DeepMind researchers Ilan Price and Matthew Wilson, it's not only better than any previous weather forecasting algorithm they have created, but even outperforms what is considered to be the most accurate system in use right now, maintained by the European Center for Medium-Range Weather Forecasts. DeepMind published its paper on GenCast in the journal Nature, where it talks about how using machine learning for weather prediction has helped it to achieve fewer errors in forecasting than traditional systems, which utilize "physics-based simulations of the atmosphere". GenCast, the researchers explained, was trained on decades worth of "reanalysis data", and is capable of generating 15-day weather forecasts in just eight minutes, compared to hours for the supercomputers running ECMWF. Naturally, DeepMind wanted to see how GenCast stacks up against ECMWF, which is the system that's used by 35 countries for their weather forecasting needs. In a series of tests that compared the 15-day weather forecasts of GenCast and ECMWF's system, DeepMind's model proved to be more accurate 97.2% of the time. With lead times greater than 36 hours, GenCast was even better, showing more accuracy 99.8% of the time. The researchers explained that GenCast is a diffusion model, based on the same kind of technology that powers its Gemini family of generative AI models. DeepMind trained the system on almost 40 years' worth of high-quality weather data curated by the European Center for Medium-Range Weather Forecasts, and the predictions it generates are said to be "probabilistic", so they account for various possibilities, expressed as percentages. AI researchers consider probabilistic models to be more nuanced than deterministic ones, which can only offer a guess at what the weather might be like on any given day. What's most impressive about GenCast is that it can get by with far less computing power than the system used by ECMWF. DeepMind said it can generate 15-day weather forecasts in just eight minutes using a single TPU v5 tensor processing unit. In comparison, the ECMWF system requires an enormous supercomputer with tens of thousands of processors to generate the same, physics-based forecast. "I'm a little bit reluctant to say it, but it's like we've made decades worth of improvements in one year," Rémi Lam, lead scientist on DeepMind's previous AI weather program, told The New York Times. "We're seeing really, really rapid progress." DeepMind said it's making GenCast open-source on GitHub, so anyone can download it along with sample code. In addition, it also plans to integrate GenCast with Google Earth, where anyone will be able to use it to generate weather predictions. As for Genie 2, this is another far-reaching development that could pave the way for developers to create expansive and playable 3D game worlds using a simple text prompt. It's said to be the next evolution of DeepMind's Generative Interactive Environments tool, which uses AI to build interactive virtual environments. The original model, Genie 1, could only create 2D worlds, but Genie 2 ventures into the 3D realm, making it far more impressive. According to DeepMind, Genie 2 is a "world model" that's able to simulate a virtual world along with animations, physics and even support interactions between all of those elements. Users can create their worlds using a prompt image to depict it, or alternatively they can first generate an image with a text prompt, then use that as the basis of the world they want to create. The possibilities are endless. Ask for a sailing simulation and Genie 2 will immediately create it. Or if you want a cyberpunk Western, it can create that too. All that's required is a reference image to start, and if there isn't one, it can even create that first. The worlds support interactions between the player's character, which can be controlled by a human or an AI agent, although at present they suffer from problems with stability. Quite simply, they're not that stable, and the model begins to lose coherency after around 20 seconds or so. DeepMind said this is partly due to Genie 2's ability to support "counterfactuals" , which are the different paths a player can take from a fixed starting point. For instance, the player can turn right or left to show a different part of the world, but the model must still take into account what's happening outside of the player's view, in case they turn back towards that scene. Still, the potential is clear, and DeepMind said it can also support various perspectives, such as an isometric, third-person or first-person view. It also depicts complex physics, such as what happens if a player or an object hits the water, how it ripples. It also showed an ability to model smoke, gravity and reflections. The environments could be useful for training AI models. In one demonstration, DeepMind's team showed how an AI-controlled character could be told to go through a specific door with a text prompt. The AI can recognize the command and apply it to the rendered environment, then immediately proceed. DeepMind didn't say much about its plans for Genie 2, such as if and when it might be released. It also refused to say what kind of computing power it requires. The lack of stability suggests there's still much work to be done, but the fact it's being worked on suggests that AI-generated gaming worlds at some point in time are a distinct possibility.

Google's AI weather forecaster, Cohere's search, and Tenstorrent raises $693 million: This week's AI launches

Google (GOOGL) DeepMind announced GenCast this week -- a high resolution AI ensemble model that can give "better forecasts of both day-to-day weather and extreme weather events" than the world's top modeling system -- the ENS from the European Centre for Medium-Range Weather Forecasts. GenCast can forecast weather events up to 15 days in advance. The diffusion model has "adapted" to the Earth's spherical geometry, and can "learn" how to more accurately generate future weather scenarios. "GenCast marks a critical advance in AI-based weather prediction that builds on our previous weather model, which was deterministic, and provided a single, best estimate of future weather," Google DeepMind said. "By contrast, a GenCast forecast comprises an ensemble of 50 or more predictions, each representing a possible weather trajectory."