Climate science, extremes, and mitigation.
Researchers developed advanced AI models that can more accurately predict individual human decision-making, which could improve strategic planning and policymaking in high-stakes situations.
Dangerous winds whipped up fast-spreading fires and dust storms across the Southern Plains, putting residents at risk from these extreme natural hazards driven by the region's dry conditions.
A new survey found that people are willing to tolerate less accurate AI in their personal lives compared to work, as they care more about convenience than precision in their day-to-day activities.
Researchers have created a global daily streamflow forecasting model that can predict flooding and water availability more accurately by combining climate data and machine learning. This can help communities better prepare for floods and manage water resources.
Researchers have used machine learning to improve computer models that track carbon monoxide levels, which can help us better understand and predict air pollution and climate change impacts.
Researchers have developed a new metric called AI-CARE to evaluate the carbon footprint of AI models, which is important as machine learning grows more energy-intensive.
Heavy snowfall and storms blanketed Japan's northern island of Hokkaido in early 2026, showcasing the region's susceptibility to extreme winter weather events that can disrupt local communities.
Two powerful cyclones caused widespread flooding in Madagascar, affecting tens of thousands of people and highlighting the growing threat of extreme weather events fueled by climate change.
Scientists developed a new model using graph neural networks to simulate how sea ice floes collide and interact, which could help improve climate predictions about the impacts of melting ice.
Researchers developed a tool to help experts collaboratively design policies for governing the behavior of large language models, which are increasingly used in high-stakes domains like mental health.
Ongoing citizens' assemblies allow for rotating citizen panels that maintain long-term public engagement on policy issues, providing a more representative and continuous voice for citizens over time.
NASA's CSDA program expanded data access agreements, giving more people the ability to use satellite imagery data for their own projects.
Researchers have been continuously measuring the Sun and Earth's energy balance for over 40 years, providing critical data for understanding and modeling our planet's climate.
Researchers developed a framework to better assess and govern risks from advanced AI systems, which are becoming increasingly autonomous and difficult to control, posing potential threats that need to be addressed.
Researchers developed advanced AI models that can more accurately predict individual human decision-making, which could improve strategic planning and policymaking in high-stakes situations.
Dangerous winds whipped up fast-spreading fires and dust storms across the Southern Plains, putting residents at risk from these extreme natural hazards driven by the region's dry conditions.
A new survey found that people are willing to tolerate less accurate AI in their personal lives compared to work, as they care more about convenience than precision in their day-to-day activities.
Researchers have created a global daily streamflow forecasting model that can predict flooding and water availability more accurately by combining climate data and machine learning. This can help communities better prepare for floods and manage water resources.
Researchers have used machine learning to improve computer models that track carbon monoxide levels, which can help us better understand and predict air pollution and climate change impacts.
Researchers have developed a new metric called AI-CARE to evaluate the carbon footprint of AI models, which is important as machine learning grows more energy-intensive.
Heavy snowfall and storms blanketed Japan's northern island of Hokkaido in early 2026, showcasing the region's susceptibility to extreme winter weather events that can disrupt local communities.
Two powerful cyclones caused widespread flooding in Madagascar, affecting tens of thousands of people and highlighting the growing threat of extreme weather events fueled by climate change.
Scientists developed a new model using graph neural networks to simulate how sea ice floes collide and interact, which could help improve climate predictions about the impacts of melting ice.
Researchers developed a tool to help experts collaboratively design policies for governing the behavior of large language models, which are increasingly used in high-stakes domains like mental health.
Ongoing citizens' assemblies allow for rotating citizen panels that maintain long-term public engagement on policy issues, providing a more representative and continuous voice for citizens over time.
NASA's CSDA program expanded data access agreements, giving more people the ability to use satellite imagery data for their own projects.
Researchers have been continuously measuring the Sun and Earth's energy balance for over 40 years, providing critical data for understanding and modeling our planet's climate.
Researchers developed a framework to better assess and govern risks from advanced AI systems, which are becoming increasingly autonomous and difficult to control, posing potential threats that need to be addressed.