Tomorrow.io's Resilience Platform is Here. Learn More.

X
Ruth Favela
By Ruth Favela
Ruth Favela
Ruth Favela
Ruth Favela is Tomorrow.io's AI Marketer. She draws on over 5 years of experience as an editor, writer, and social media manager for AI startups, B2B SaaS, and B2C products. In her role, Ruth focuses on using AI tools to create customer-first content for the various industries Tomorrow.io has solutions for. She writes about weather innovations, AI/ML modeling, weather API applications, weather AI use cases, and much more.
Jan 17, 2024· 4 min, 23 sec

ClimaCon 4 Keynote: Generative AI at Tomorrow.io

    TL;DR:

    • Tomorrow.io is using AI to transform weather forecasting and achieve breakthrough gains in predictive accuracy.
    • Vast data sets and machine learning are training superior AI models for Tomorrow.io.
    • AI is driving exponential leaps in forecast accuracy, with reductions in key global forecast error rates.
    • Probabilistic forecasting enables superior risk management compared to traditional deterministic forecasts.
    • AI assistants like Gale translate weather insights into actionable information for decision makers.

    AI took the world by storm in 2023. New applications and advancements arose across every single industry.

    At ClimaCon 4, Tomorrow.io’s annual weather intelligence and climate adaptation conference, company leaders revealed how artificial intelligence is transforming weather forecasting, enabling breakthrough gains in predictive accuracy.

    This blog will recap the exciting update shared by Tomorrow.io’s Director of Data Science, Tyler McCandless, and VP of Growth & Strategy, Cole Swain.

    Vast Data Sets Train Tomorrow.io’s Superior AI Models

    “Machine learning fills a lot of those gaps. And then, kind of at the end of it, we want to have better probabilistic forecast,” explained Tyler McCandless, Tomorrow.io’s Director of Data Science. He emphasized that the quality and breadth of training data is critical for AI models.

    Tomorrow.io’s vast proprietary datasets derived from satellites, radars, IoT sensors and numerical weather prediction models provide unmatched training power.

    “Now we’re getting better and better observations updated faster. We can use machine learning to learn that relationship between how the numerical weather prediction models expect the atmosphere to evolve and what’s actually occurring,” McCandless said.

    AI Drives Exponential Leaps in Forecast Accuracy

    McCandless revealed that Tomorrow.io is achieving stunning 30-60% reductions in key global forecast error rates like temperature and precipitation. “That’s a big advancement and the ability to use the forecast for better decisions,” he remarked.

    Cole Swain, VP of Growth & Strategy, added that Tomorrow.io is seeing “measurable improvements and dramatic changes with what Tyler and team and all those around are really uncovering.” He emphasized that “what we talk about today is really just gonna be the beginning of all that we learn over the course of the next few weeks, next few months, next few years.”

    What we’re doing today with AI and modeling will set the stage for years to come.

    Probabilistic Forecasting Enables Superior Risk Management

    McCandless explained how probabilistic forecasting provides businesses and governments superior risk management capabilities compared to traditional deterministic forecasts.

    “Seventy percent chance of precipitation matters a whole lot differently if you’re in the ski resort industry compared to, you know, if you’re an event company,” he said. Probabilistic forecasts calibrate the likelihood of events, allowing users to match decisions to their risk tolerance.

    Faster Model Runs Enable More Timely Predictions

    McCandless noted that artificial intelligence enables Tomorrow.io to run forecast models exponentially faster than previous generations of technology allowed. This increased speed translates into more timely delivery of predictions to users.

    “When you think about global weather models being updated every 6 hours, you miss a lot of how the forecast has evolved in that time period.. So now with machine learning, we can ingest those, improve the prediction, adding probabilities within that time period, or you can capture when the weather’s rapidly changing,” he explained.

    By updating forecasts every hour or even every few minutes, Tomorrow.io provides users an information advantage to proactively respond to changing conditions.

    AI Assistants Translate Insights for Decision Makers

    Tomorrow.io is leveraging natural language processing in tools like its AI assistant Gale. Gale digests reams of forecast data and conveys tailored insights in natural language.

    As Swain described, “Instead of, say, an executive or any type of dispatcher or manager having to look at a dashboard and piece together this whole story, Gale understands how to create a snapshot. So it’s just a paragraph and you’d be so surprised by the amount of power of information you can carry through text.” And that’s the power of weather AI.

    Interested in Gale? Learn more about Gale and weather AI today. 

    A Bright Future for AI in Weather Modeling & Forecasting

    Tomorrow.io is pioneering the use of AI in weather forecasting, but company leaders say they have only scratched the surface of this technology’s potential.

    As Swain remarked, “What we talk about today is really just gonna be the beginning of all that we learn over the course of the next few weeks, next few months, next few years.

    We’re at the tip of the iceberg here.”

    This keynote revealed how Tomorrow.io is leveraging artificial intelligence to achieve revolutionary gains in the accuracy and timeliness of weather prediction.

    AI assistants like Gale are translating these predictive superpowers into real-world insights that enable better decision making. There is no doubt

    And with the data coming from Tomorrow.io’s proprietary satellite constellation, these models will be fueled with the most robust information available.

    Learn more about Tomorrow.io advancements in AI and ML and weather modeling at ClimaCon4.

    See Tomorrow.io in Action!

    This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

    TOS(Required)

    Learn more about Tomorrow.io advancements in AI and ML and weather modeling at ClimaCon4. 

    Exit mobile version