AI could warn people of dangerous weather patterns
Add Axios as your preferred source to
see more of our stories on Google.
Illustration: Maura Losch/Axios
Climate tech company Tomorrow.io is building a new generation of weather satellites that could give billions of people earlier warnings about dangerous storms.
Why it matters: It's an example of the kind of tangible upside that can be hard to spot amid escalating warnings about AI's risks to humanity.
Driving the news: Tomorrow.io on Thursday unveiled the full instrument architecture for DeepSky, its next-generation early-warning satellite constellation.
- The company says more than 5 billion people live outside ground-based weather radar coverage, with large blind spots over oceans where storms can intensify before reaching populated areas.
- That lack of coverage leaves these communities vulnerable to unpredictable and potentially catastrophic weather events.
How it works: DeepSky is essentially a data-gathering layer designed to give AI forecasting models better real-time data.
- It has satellites that carry five instruments that measure different parts of the atmosphere: temperature, moisture, ocean-surface winds, cloud structure and precipitation.
- Those satellites are purposely deployed to cover oceans and other places with little or no radar coverage.
- Those observations will be fed into numerical and AI weather models.
Zoom out: AI has improved the capabilities of weather modeling.
- AI-powered cameras, satellite systems and weather-prediction tools are helping agencies in Western states detect wildfires sooner, for example.
- But "AI is only as powerful as the data behind it," Tomorrow.io CEO and co-founder Shimon Elkabetz told Axios via message.
Yes, but: Tomorrow.io has claimed it would use AI tools to fix weather forecasting for underserved communities for years.
- The scale of the improvement in real-world warning times remains to be proven upon further rollout of the technology.
