Google DeepMind has released an AI weather forecasting model that updates every hour using satellite imagery. The system predicts wind speeds at turbine height and solar irradiance for power markets.
DeepMind's latest model addresses a critical gap in renewable energy forecasting. By refreshing predictions hourly rather than at longer intervals, the system provides grid operators with more granular data for managing fluctuating wind and solar output.
The model processes satellite imagery to forecast conditions specifically relevant to power generation—wind velocity at heights where turbines operate and sunlight intensity reaching solar installations. This precision helps utilities optimize energy dispatch and balance supply across their networks.
Hourly updates represent a significant improvement over traditional weather forecasting cycles. Grid operators can now make faster, more informed decisions about renewable energy integration and backup power requirements.
The development reflects growing focus on AI applications in climate tech and energy infrastructure. Accurate short-term renewable forecasting reduces reliance on fossil fuel backup systems and improves the efficiency of distributed power networks.
Family-focused AI assistant Ollie is positioning privacy as a competitive advantage, promising not to use user data for model training or third-party sharing.
NVIDIA has released PAIR, a free open-source tool that distributes AI workloads across idle computers. The platform enables organizations to leverage unused processing power for machine learning tasks.
OpenAI confirmed a major outage affecting ChatGPT and Codex, with users reporting widespread errors across core features. The downtime comes as the company prepares to launch its new Astra model.
Anthropic's Claude AI service experienced elevated error rates across multiple models, according to a status page incident report. The issue drew significant attention on Hacker News with 75 comments and 105 points.