×
Google to launch 50+ AI satellites for 15-minute wildfire detection
Written by
Published on
Join our daily newsletter for breaking news, product launches and deals, research breakdowns, and other industry-leading AI coverage
Join Now

Google has partnered with nonprofit Earth Fire Alliance and satellite manufacturer Muon Space to launch Fire Sat, a constellation of specialized wildfire-tracking satellites that will use AI to detect fires in their earliest stages. The initiative aims to deploy 50+ satellites by 2029, capturing images of fire-prone areas every 15 minutes and processing the data through Google’s machine learning systems to distinguish actual fires from false positives like hot roofs or reflected sunlight.

What you should know: Fire Sat represents Google’s ambitious entry into space-based wildfire monitoring, combining satellite technology with AI processing to transform early fire detection.

  • The first satellite launched in March 2025, with three more planned for early 2026 and the full 52-satellite constellation expected by 2029.
  • At full capacity, the system will detect fires as small as ten square meters and provide updates every 15 minutes.
  • Google plans to showcase images from the current test satellite sometime this summer.

How it works: Fire Sat satellites use dual-camera systems and advanced AI processing to overcome the limitations of current space-based fire detection methods.

  • Each satellite captures images using both standard visible/short-wave infrared cameras and cryo-cooled thermal cameras that provide higher resolution than traditional microbolometer sensors (heat-detecting chips that don’t require cooling).
  • The dual imagery is sent to Google data centers where computer vision and machine learning analyze the photos, cross-referencing different image types and comparing against historical data.
  • “The whole job of the constellation after it collects the data is really to funnel it to a data center where we can take the imagery and analyze it to understand if what we’re looking at is likely a fire or a false positive,” says Christopher Van Arsdale, a Google researcher.

Why this matters: Current satellite systems struggle with early fire detection due to grainy, low-resolution thermal imaging that creates numerous false positives.

  • “If you look at a noisy picture, everything kind of looks like a tiny fire,” Van Arsdale explains. “You need these very high-fidelity pictures in order to actually do a good job with detection.”
  • The 15-minute update window aims to catch small fires before they grow too large, providing actionable intelligence for first responders.
  • “With fire in particular, times are compressed so much that you have to apply technology to make a decision within the timeframe that you can impact the outcome of what’s happening,” says Brian Collins, executive director of Earth Fire Alliance.

Competitive landscape: Google isn’t alone in pursuing space-based wildfire monitoring, but it’s moving faster than government-funded alternatives.

  • Canada’s WildfireSat program plans to launch fire-specific satellites in 2029, the same timeline as Fire Sat’s full deployment.
  • Nearly 9 million acres have already burned in Canada’s 2025 fire season, highlighting the urgent need for better monitoring systems.
  • Existing ground-based camera networks like AlertWildfire have successfully spotted fires, including the deadly Palisades Fire in Los Angeles.

The bigger picture: Fire Sat emerges amid increased private sector investment in wildfire technology and reduced federal disaster response funding.

  • President Trump’s June executive order called for prioritizing fire tech companies while combining federal disaster agencies and declassifying satellite data.
  • Sweeping cuts to FEMA and the US Forest Service mean private industry is moving to fill gaps in public disaster response capabilities.
  • Google’s 2024 emissions increased 50% due to generative AI efforts, making Fire Sat potentially a form of environmental atonement.

What experts are saying: While the technology shows promise, concerns remain about data accessibility and long-term sustainability of private solutions.

  • “The trend towards AI to assist with all of this is obviously going to produce better results, but it’s not going to produce consistent results necessarily,” says Krystal Azelton of the Secure World Foundation, a nonprofit that advocates for sustainable space policies.
  • Climate scientist Daniel Swain notes: “It doesn’t really solve the core underlying problems, but it’s probably a beneficial thing to do. It unfortunately doesn’t give us much of an edge under the most extreme conditions.”
  • “We’ve seen this happen” with tech companies discontinuing services, Swain warns, pointing to Google’s history of killing off products.
Google Wants to Get Better at Spotting Wildfires From Space

Recent News

NiCE rebrands as human-centered AI platform to augment customer service

New CEO Scott Russell steers away from automation hype toward measurable outcomes.

We tested 3 AI travel agents and here’s what actually works

Ideal for brainstorming trips, but you'll still need to handle the actual bookings.

Salesforce unveils Agentforce 3 with unified studio and monitoring tools

Real-time monitoring features could finally give enterprises confidence to deploy AI agents with customers.