Cloudy with a chance of algorithms

Summary

---

Click Here: "Cloudy with a Chance of Algorithms"

Podcast: Click Here (Recorded Future News)
Host: Dena Temple Raston
Guest: Paris Perdicaris (Associate Professor, University of Pennsylvania)
Release Date: September 26, 2025
Summary by Section, with Quotes and Timestamps

---

Overview

This episode delves into the evolving role of artificial intelligence in weather forecasting. Host Dena Temple Raston and guest Paris Perdicaris discuss how AI could revolutionize predictions, surpass traditional meteorology in speed, scope, and accuracy, and even democratize access. They also confront the limitations and risks—including the impact of data loss due to political and funding changes—and emphasize that while AI offers promise, its accuracy and reliability still depend on high-quality human-provided data and expertise.

---

Key Discussion Points & Insights

The Promise—and Limits—of AI Forecasts

• AI as an Enhancement, Not a Replacement
  • Paris addresses the common fear that AI will replace human forecasters, referencing Hollywood's "Terminator" as a metaphor for public anxiety:

    "The general fear that those systems are kind of going to be like a Terminator movie and kind of replace humans and so forth. At least in the field of Earth system science and weather forecasting, they're just going to enhance our ability to better predict the Earth system rather than replacing a human expert."
    (Paris Perdicaris, 00:54)

  • Dena jokes, “So no killer robots, just better predictions.” (01:20)
  • AI's goal: Make forecasting tools accessible to everyone, “so people can run their own forecasts right on their laptops.”
    (Dena Temple Raston summarizing, 01:20-01:45; Paris Perdicaris, 01:37)

The Data Behind the Forecast

• Vast Historical Records Enable AI Progress

  • Data from satellites, weather stations, balloons, aircraft, and ships has accumulated since the 1950s, forming an “information goldmine.”
    (Dena Temple Raston, 03:27)
  • Paris' AI system, Aurora, was trained on this wealth of data.

• Aurora: How AI ‘Learns' the Weather

  • Aurora “doesn’t understand the atmosphere” in a human sense; rather, it recognizes and learns patterns:

    "The way a sailor predicts the weather is experience based, right? ...They kind of feel the wind and see the clouds, and from that information, they can make a prediction of what's about to happen."
    (Paris Perdicaris, 03:47)

  • Aurora, unlike a sailor, “just sees it in terms of patterns.”
    (Dena Temple Raston, 04:00)

• Strengths and Limitations

  • Aurora “outperforms all existing operational forecasts for hurricane tracking,” providing accurate tracks four to five days before land impact.
    (Paris Perdicaris, 04:06)
  • However, critics cite AI’s struggle with rapid, small-scale events like flash floods or microbursts, which are increasingly relevant.
    (Dena Temple Raston, 04:25)

Building a Smarter Model

• Going Beyond Weather: Integrating Diverse Data
  • Paris includes “data from the oceans, air chemistry and air pollution...things like fossil fuel emissions, ocean chemistry, air pollution, concentration of carbon dioxide in the atmosphere.” (Paris Perdicaris, 05:08, 05:26)
  • The model's training mimics human education: a “general education phase” followed by advanced specialization (e.g., hurricanes, air quality).
    (Paris Perdicaris and Dena Temple Raston, 05:38–06:15)

AI vs. Traditional Meteorology

• How AI Delivers Faster, Cheaper Forecasts

  • “AI models excel by learning patterns in actual data, rather than trying to solve complicated systems of mathematical or physics equations like the traditional tools are doing.”
    (Paris Perdicaris, 08:04)
  • Aurora can deliver forecasts from 6 hours to 10 days ahead.
    (Paris Perdicaris, 08:21)

• Human Expertise Remains Crucial

  • “If I give my model the wrong inputs and then it tells me that it’s going to snow in Greece in August, then obviously that is an erroneous prediction that I should filter out.”
    (Paris Perdicaris, 08:33)
  • Final forecasts still depend on “a human meteorologist...has synthesized many different forecasts...and based on their expertise, are issuing the official forecast.”
    (Paris Perdicaris, 08:47)

The Politics and Perils of Data

• AI Needs Data—Quality and Quantity

  • Funding and staffing cuts to NOAA (National Oceanic and Atmospheric Administration) threaten the “raw material” of AI models.

    "If those changes impact the quality of the data we have access to and the quality at which we are able to monitor the Earth system, that's going to have a direct impact on the AI models, which are only as good as the data that they've seen."
    (Paris Perdicaris, 10:11)

  • Paris expresses worry about the consequences, especially given political actions aiming to privatize or dismantle NOAA.

    “The funding cuts at NOAA shook up the community...if it continues in this direction, it is going to slow down progress.”
    (Paris Perdicaris, 10:35)

• The Risk of Stale Training Data

  • The first version of Aurora was trained on comprehensive data before major cuts, but future models could struggle as climate change accelerates and “less or outdated data is the last thing forecasting needs.” (Dena Temple Raston, 10:54)

Noteworthy Optimism Amid Uncertainty

• Paris closes with a note of hope, stressing there’s “still plenty of untapped data” for the next generation of AI models:

  "We're at the beginning and in fact, we have only scratched the surface when it comes to the data that is available out there for building those AI systems."
  (Paris Perdicaris, 11:14)

Final Reflection

• Dena sums up the episode with a reminder that AI’s real test will be how well it predicts rare or severe events:

  “The real test of AI won’t be the sunny day it gets right. It’ll be the storm it never saw coming.”
  (Dena Temple Raston, 11:23)

---

Timestamps for Important Segments

• 00:54 – Paris on AI as a tool, not a threat: “No Terminator movie for weather.”
• 03:12-03:47 – The scale and source of weather data, and how Aurora ‘learns.’
• 04:06 – Aurora’s strengths (hurricane tracking) and its current limitations.
• 05:08-05:32 – Incorporating emissions, chemistry, and diverse data into forecasts.
• 08:04 – AI’s efficiency advantage over traditional meteorological models.
• 08:47 – The enduring importance of human expertise in the forecasting process.
• 09:26-10:35 – Implications of NOAA cuts and political influences on data.
• 11:14 – Paris’s optimism about future data and AI’s potential.

---

Notable Quotes

• “We want to make those tools accessible to everyone around the world that can operate them on their own personal computers.”
  —Paris Perdicaris (01:37)

• “AI models excel by learning patterns in actual data, rather than trying to solve complicated systems of mathematical or physics equations like the traditional tools are doing.”
  —Paris Perdicaris (08:04)

• “The real test of AI won’t be the sunny day it gets right. It’ll be the storm it never saw coming.”
  —Dena Temple Raston (11:23)

---

Tone & Style

• The episode strikes a conversational, approachable tone. Technical concepts are made accessible, with references to pop culture (like “Terminator”) and relatable analogies (sailors, liberal arts education).
• The guest is optimistic but realistic about AI’s limitations and dependencies.

---

Summary

This episode paints a nuanced, accessible picture of AI’s transformative potential in weather forecasting. While AI models like Aurora can process vast datasets and vastly improve forecast speed and accuracy, their power is ultimately constrained by the quality and quantity of data available—much of it threatened by political actions and budget cuts. Human expertise remains integral, both to correct AI's odd errors and to provide regional guidance. AI’s future in meteorology is bright but depends not only on algorithms, but on sustaining the institutions that supply them with data. The real proof will come when AI faces unprecedented, extreme weather events—“the storm it never saw coming.”

---

Transcript

[Paris Perdicaris]

Chatgpt AI Machine Satellite Engine Ignition.

[Dena Temple Raston]

Click here and lift up. From recorded Future News and prx, this is Click Here's Mic Drop. A longer listen to one of our favorite interviews of the week. I'm Dena Temple Raston. We all know weather forecasts are fallible. Your phone promises sunshine, and then you get rain. So what if computers could actually beat even the best meteorologists? That's the promise of AI weather forecasting, and it comes with all the usual hype. Paris Perdicaris is an associate professor at the University of Pennsylvania, and Microsoft tapped him to build a better forecast using AI. When I asked him what people get wrong about that, he went straight to the Hollywood metaphor.

[Paris Perdicaris]

The general fear that those systems are kind of going to be like a Terminator movie and kind of replace humans and so forth. At least in the field of Earth system science and weather forecasting, they're just going to enhance our ability to better predict the Earth system rather than replacing a human expert.

[Dena Temple Raston]

Great, so you don't actually see a Terminator movie, you know, in the future?

[Paris Perdicaris]

Not when it comes to air system prediction and weather forecasting.

[Dena Temple Raston]

Okay, so no killer robots, just better predictions. But better doesn't mean perfect. Paris says he sees a day when those tools could put a kind of digital meteorologist in every home. People can run their own forecasts right on their laptops.

[Paris Perdicaris]

We want to make those tools accessible to everyone around the world that can operate them on their own personal computers.

[Dena Temple Raston]

That sounds democratizing, but it also raises questions like what happens when the models are wrong? Who's accountable? Stay with us. Come with me if you want to live.

[Recorded Future News Announcer]

If you're looking for a daily guide to cybersecurity news and policy, sign up for the Cyber Daily from Recorded Future News. It serves up the day's most interesting and important cyber stories from our sister publication, the Record, and then aggregates all of the big cyber stories you might have missed from news outlets around the world. Just go to the Record Media and click on Cyber Daily to get all you need to know about the world of cybersecurity right in your inbox.

[Dena Temple Raston]

I'm Dena Temple Roster, and this is Click Here's Mic Drop. Paris Perdicaris has spent the past decade figuring out how AI interprets the physical world. Hurricanes, ocean currents, weather systems.

[Paris Perdicaris]

My background over the last 10 years, I've done a lot of work on AI and deep learning applied specifically to physical systems.

[Dena Temple Raston]

So it made sense that Microsoft would come calling as they were trying to figure out if AI could handle one of the Biggest data problems we have weather. Weather, it turns out, produces a ton of data.

[Paris Perdicaris]

So Starting from the 50s, 60s and 70s, we have been collecting around the clock from satellites, from weather stations, air balloons, aircrafts, ships in the oceans, or buoys.

[Dena Temple Raston]

So it's an information goldmine. Decades of public data gathered by agencies like NOAA or its European equivalent, the ecmfw. Big tech's breakthroughs depend on their treasure chest of data. So Paris built Aurora, an AI system that learns from all that weather data.

[Paris Perdicaris]

The way a sailor predicts the weather is experience based, right? So they're looking back to their prior experience. They've been in the situation many times. They kind of feel the wind and see the clouds, and from that information, they can make a prediction of what's about to happen.

[Dena Temple Raston]

Aurora doesn't understand the atmosphere. It just sees it in terms of patterns.

[Paris Perdicaris]

And so far we have seen that on average, it outperforms all existing operational forecasts for hurricane tracking and can generate tracks, can predict accurately tracks maybe four to five days before those have an impact on land.

[Dena Temple Raston]

Now, that sounds impressive, but there's a caveat. It appears to do well with these broad predictions like hurricanes. But critics say that broad accuracy is one thing and being able to spot the conditions that lead to, say, flash floods and microbursts is another. And those are the storms that seem to show up more often these days, so they've started to matter more. Microsoft isn't the only big tech player trying to use AI to predict weather. Google's been developing AI weather tools. And Huawei in China, it has too. They all feed on the same public data. Paris also added a few other data points to his model.

[Paris Perdicaris]

We tried to go a step beyond that and use a lot more diverse data, include also data from the oceans, you know, air chemistry and air pollution. To build a more general model that goes beyond weather forecasting, he added things.

[Dena Temple Raston]

Like fossil fuel emissions, ocean chemistry, air.

[Paris Perdicaris]

Pollution, for instance, concentration of carbon dioxide in the atmosphere or concentration of nitrogen dioxide.

[Dena Temple Raston]

Feed the model millions of examples, and slowly it begins to connect the dots.

[Paris Perdicaris]

It's a little bit like your brain when you're learning. So the first phase you can think of as a general education phase, where the model is going to see through all these decades of historical data. And now we're talking about millions of hours of, you know, Earth evolution data, everything from temperatures and wind patterns and ocean currents.

[Dena Temple Raston]

It's like giving an AI system a liberal arts education in Earth systems. Then you send it to graduate school to learn about hurricanes, air quality, and.

[Paris Perdicaris]

Ocean waves, for instance, high resolution weather forecasting, or air quality prediction, or ocean wave prediction, or hurricane tracking, or any other interesting prediction task that we might be interested in in order to understand and monitor the Earth system.

[Dena Temple Raston]

But forecasts don't live in a vacuum. They depend on humans and politics. That's when we come back. Stay with us.

[Recorded Future Advertisement Narrator]

In cybersecurity, your greatest fear isn't the threats you see. It's the critical signals lost in the noise. Every day, security teams sort through millions of potential threats. That's why Recorded Future exists, to give you precision intelligence tuned to your needs. Our advanced AI detects patterns humans might miss, while our threat intelligence experts, veterans of military and intelligence services, provide crucial context. With Recorded Future, you gain the confidence to identify critical threats and the precision to act before they become attacks. Learn why 1900 + customers, including 45 + sovereign governments, trust us to detect threats faster and and achieved 350% plus ROI within a year.

[Dena Temple Raston]

Traditional weather prediction takes labor, balloons, equations, supercomputers. What AI can do is shortcut the math.

[Paris Perdicaris]

AI models excel by learning patterns in actual data, rather than trying to solve complicated systems of mathematical or physics equations like the traditional tools are doing.

[Dena Temple Raston]

That makes them faster and cheaper and sometimes more accurate.

[Paris Perdicaris]

So tools like Aurora generate everyday forecasts basically from six hours ahead, maybe up to 10 days ahead.

[Dena Temple Raston]

But here's the garbage in, garbage out.

[Paris Perdicaris]

If I give my model the wrong inputs and then it tells me that it's going to snow in Greece in August, then obviously that is an erroneous prediction that I should filter out.

[Dena Temple Raston]

And that's where humans come in.

[Paris Perdicaris]

There's an additional element here and layer, which is the human expertise of a human meteorologist. So the way official forecasts are issued is through a human that has a specific expertise for the region. That human has synthesized many different forecasts from Aurora and other models and based on their expertise, are issuing the official forecast.

[Dena Temple Raston]

So AI isn't replacing meteorologists, at least not yet. But what if the data itself begins to disappear?

[Recorded Future Advertisement Narrator]

Hundreds of employees at the national oceanic and Atmospheric Administration have been fired, and more layoffs in the agency could be coming today.

[Dena Temple Raston]

In the early days of the second Trump administration, some 1300 members of Noa either resigned or were laid off, and the agency still faces more cuts. And as we reported On Tuesday, Project 2025, the administration's legislative playbook, calls for NOAA to be disbanded and privatized. And that could mean fewer weather balloons, fewer satellites, less data. Since so much of AI is, you know, good information, good Data producing good results. If you have, say, half the number of balloon launches or half the number of satellites actually looking at the weather, is that going to inevitably affect the ability for Aurora to be accurate?

[Paris Perdicaris]

Absolutely, yes. I mean, likely. If those changes impact the quality of the data we have access to and the quality at which we are able to monitor the Earth system, that's going to have a direct impact on the AI models, which are only as good as the data that they've seen.

[Dena Temple Raston]

When Paris heard about the NOAA cuts, he says he immediately worried for Aurora.

[Paris Perdicaris]

Yeah, absolutely. I mean, the funding cuts at NOAA shook up the community and definitely for us at develop AI models, we are relying on noaa. We run relying on their expertise, their data. So of course it's a blow. And you know, if it continues in this direction, it is going to slow down progress.

[Dena Temple Raston]

The first version of Aurora was trained before the cuts. It runs fine, at least for now, but in a few years it'll need new data. And with the climate shifting faster than ever, less or outdated data is the last thing forecasting needs. Though Paris, for his part, remains optimistic.

[Paris Perdicaris]

We're at the beginning and in fact, we have only scratched the surface when it comes to the data that is available out there for building those AI systems.

[Dena Temple Raston]

Optimistic. There's still plenty of untapped data. But the real test of AI won't be the sunny day it gets right. It'll be the storm it never saw coming. From Recorded Future News. This has been Click here's mic drop. It was written and produced by Megan Dietry, Sean Powers, Erica Gaeda, Zach Hirsch, Lucas Riley and me, Dina Temple Rousten. It was edited by Karen Duffin. We'll be back on Tuesday with an all new episode of Click Here. Have a great weekend. SA.

[Recorded Future News Announcer]

Looking for more of the cybersecurity and intelligence coverage you get on Click Here. Then check out our sister publication, the Record. From Recorded Future News. You'll get breaking cyber news from reporters in New York, Washington, London and Kyiv, among others. And you'll see for yourself why it attracts hundreds of thousands of page views every month. Just go to the Record Media.