Get updates delivered to you daily. Free and customizable.
Interesting Engineering
How one startup is using AI to cool a warming world
By Bojan Stojkovski,
1 day ago
This summer, the tiny Balkan country of North Macedonia has faced unusually high temperatures, causing various challenges across the country. Firefighters have worked to contain multiple wildfires fueled by strong winds and dry conditions, while smaller fires have also been reported in Greece and other parts of the Balkans.
With temperatures exceeding 40 degrees Celsius (104°F) in some areas, authorities advised people to stay indoors and limit outdoor activities during the hottest parts of the day. Extreme summer temperatures in Europe have risen significantly faster than the global average.
According to a recent analysis published in the New York Times , the average person has encountered 26 additional days of unusual warmth due to global warming since last May .
“My home city of Skopje will be hit by summers that will be hotter by 8℃ if we don’t control GHG emissions, summers that already peak at +40℃!”, says homegrown eco-activist and entrepreneur Gorjan Jovanovski.
For the young Macedonian, tackling climate change turned into a lifelong mission. He first started through AirCare, a mobile app designed to map and visualize air pollution in the Balkans. The app leveraged open data from government and volunteer air quality sensors to provide users with up-to-date information on environmental conditions.
Jovanovski’s latest startup , EarthCare.ai, provides companies with tools and data to understand and anticipate natural hazards like floods, wildfires, and storms. By using artificial intelligence for accurate and timely environmental predictions, the startup helps businesses assess risks and prepare for various environmental challenges.
“I honestly believe that the biggest challenge and threat of our generation is saving the planet. We already know that we’re deep in the climate crisis, and it will take multiple solutions to reverse what we’ve done to our planet,” he tells IE.
Not doing anything will cost businesses more
The idea behind EarthCare.ai originated from a deep concern about the climate crisis, which Jovanovski sees as a major challenge of our time. After the success of AirCare, which made air quality data accessible, Jovanovski and his co-founder and CTO, Dragan Gelevski, sought to broaden this approach to address a wider spectrum of climate and weather issues.
The impact of 50 years of extreme weather events, intensified by human-induced global warming, has resulted in over two million deaths and $4.3 trillion in economic losses.
Hence, businesses are increasingly exposed to climate-related risks. Examining these emerging threats and their financial repercussions, Jovanovski, drawing on his background in software engineering, recognized that without clear, actionable climate data, companies would encounter escalating challenges and costs.
“Businesses not only are going to be really impacted by the new climate reality, not doing anything about it will cost them more. Hence, providing understandable and high quality climate data allows businesses to better plan and mitigate upcoming storms, floods, droughts, wildfires and more,” Jovanovski explains.
According to him, EarthCare.ai aims to make climate data as accessible as Excel-made numerical data.
“Energy companies have massive infrastructure monitoring issues that can be affected by storms (imagine solar farms getting hit by hail). Insurances need to better price policies based on the new reality of severe weather. Banks giving out loans need to better risk-assess if that new factory will get flooded or not. The list goes on, since climate and physical risk affects so many industries around the world, even if they aren’t yet aware of it. Getting a head start gives so many benefits, rather than waiting to start losing money first,” Jovanovski explains.
Improving climate data resolution with radar and ground monitoring
While many climate platforms rely solely on satellite data, Jovanovski incorporates multiple sources, including radar systems and ground monitoring.
“This not only gives us continuous data, but also lets us have a much better resolution and granularity. From the model side, we’ve started developing our own threat detection models that are able to transform meteorological data into actual detected threats (like floods, storms, droughts, wildfires…), as well as putting together the latest AI forecast models to see well into the future,” Jovanovski adds.
As AI is still developing , advanced models such as LLMs and GenAI represent the early stages of its potential. Jovanovski argues that as climate models advance, they promise to provide a more precise understanding of environmental changes and offer scenarios to address and mitigate climate impacts.
“There are already AI climate models that can create accurate forecasts for the next 10 days in minutes on a laptop, compared to the conventional physics-based models that take hours running on supercomputers to achieve the same. Yes, AI draws a lot of power, and will set us back at the start in terms of carbon, but when used for the right things, it will give us an enormous ROI and boost forward,” he adds.
For the young startup, the ultimate goal is to create an AI model that serves as a digital twin of our world. The next step involves running simulations to determine the best actions for addressing climate challenges.
“The start is having all of the world’s climate data centralized in one place, and that is how we’ve started to tackle the problem. We have to do it now because it is the last chance we have to let our kids grow up in a world that isn’t dry, on fire or under water. And with AI, there is no better weapon out there to achieve this goal.” Jovanovski concludes.
Get updates delivered to you daily. Free and customizable.
It’s essential to note our commitment to transparency:
Our Terms of Use acknowledge that our services may not always be error-free, and our Community Standards emphasize our discretion in enforcing policies. As a platform hosting over 100,000 pieces of content published daily, we cannot pre-vet content, but we strive to foster a dynamic environment for free expression and robust discourse through safety guardrails of human and AI moderation.
Comments / 0