Improving the Physical Realism of AI Weather Forecasts: Scott Hosking at AI for Science @TheAlanTuringInstituteUK
Improving the Physical Realism of AI Weather Forecasts: Scott Hosking at AI for Science  @TheAlanTuringInstituteUK
Uploaded May 2026 | Updated November 2026, 3 hours ago
Messy Data, Fastnet and Resilient Forecasting: Scott Hosking at AI for Science

At the AI for Science event at the Royal Society, Scott Hosking (Alan Turing Institute) discusses how “messy” environmental observations must be made AI ready to improve forecasting and resilience, set against rising costs and impacts from extreme weather and related national security risks. He introduces the UK Met Office partnership on Fastnet, an AI weather forecasting model trained on ERA5, explains its graph neural network and multimesh design, and describes work to reduce smoothing and grid artefacts while improving physical consistency without harming skill scores. He also presents a data sparse region approach that learns directly from fragmented observations for Sub-Saharan Africa and outlines an 18 month pilot with local agencies. Hosking closes with progress toward longer range climate prediction, use of open sourced foundation models, UK supercomputing, and the importance of interdisciplinary teams and project management.

Scott Hosking is Mission Director for Environmental Forecasting at the Turing Institute. In 2018, he established the BAS AI Lab and formed a research team focused on Arctic sea ice forecasting. Working in collaboration with Fellows at the Turing, he developed the widely recognised IceNet programme, which delivered the first AI-powered pan-Arctic sea ice forecasting model.

00:00 Session Kickoff
00:16 Meet Scott Hosking
01:02 Messy Data Challenge
01:27 Extreme Events Rising
04:56 Climate as Security Risk
06:15 Turing Work Overview
07:53 AI for Science Pillars
09:13 FastNet Weather Forecasting
11:26 ERA5 Training Data
12:42 FastNet Architecture
14:58 RMSE Smoothing Limits
17:15 Artifacts and Fixes
22:23 Forecasting Data Sparse Regions
23:10 Learning from Raw Observations
27:13 Operationalizing FastNet
29:53 Toward Climate Prediction
32:14 Foundation Models Ecosystem
34:07 People and Culture Template
36:18 Thanks and Wrap Up
Improving the Physical Realism of AI Weather Forecasts: Scott Hosking at AI for ScienceHongbo Zhao - Learning Battery Physics from ImagesHow to successfully deploy an AI workflow in your business?Perspectives on practical reproducibility in climate scienceCleared for Takeoff: Navigating the Future of AI Air Traffic ControlLiving with Machines: The digitisation processThe Cyber Cold War: How AI is Defending Our Critical National InfrastructureWhy work in Data Science and AI?The cutting edge of emotional and cognitive neuroscienceLiving with Machines: Computational methods, infrastructure and skillsAIM RSF: Multimorbidity and place (Alan Marshall)CETaS Webinar: Privacy-preserving Moderation of Illegal Online Content
The Alan Turing Institute |

Improving the Physical Realism of AI Weather Forecasts: Scott Hosking at AI for Science

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER