This week’s roundup of data news highlights from July 18, 2026, to July 24, 2026, features an AI system that detects wildfires before they spread and robots that can tell the difference between objects that are safe to bump into and those that are not.
1. Tracking Songbird Evolution
Researchers at the University of Michigan have used an AI tool to study the evolution of birds, proving that birds in the group Passeriformes evolved in rapid bursts coinciding with climactic shifts throughout Earth’s history. These findings support the evolutionary theory of adaptive radiation, in which a single ancestral species rapidly diversifies into many new species due to an ecological catalyst.
2. Remotely Flying Fighter Jets
The Defense Advanced Research Projects Agency (DARPA) and the U.S. Air Force have used an AI program to successfully fly a modified F-16 fighter aircraft remotely. This is the first time the U.S. military has used a retrofitted older aircraft for AI-powered flight. The program flies the aircraft and monitors sensors and other mission systems, with a human pilot conducting takeoff and overseeing the test.
3. Improving Road Safety
A researcher at the University of Houston in Texas is using large language models to analyze roadway-condition data, including pavement structure, surface condition, roadway geometry, and crash records. The new research shows how pavement conditions influence crash occurrence and severity, identifying roadway segments in the greatest need of maintenance and safety improvements.
4. Detecting Wildfires Early
The North Summit Fire District in Utah is utilizing an AI-powered wildfire-detection camera system to help with early fire detection in remote areas. The system detects smoke, contacts a human to confirm its findings, and then sends an alert to local fire agencies. In one instance, the system detected a fire that had only spread to a single tree.
5. Navigating Cluttered Environments
Researchers at the University of South Carolina have developed an AI system that helps robots distinguish between harmless contact and risky collisions in complex environments. Whereas many robots operate on collision-free models, avoiding all contact, the researchers’ model teaches robots to avoid risky collisions, such as knocking over a full glass of water, but not harmless ones, such as knocking over an empty plastic water bottle.
6. Coordinating Military Missions
The Singapore Armed Forces’ Digital and Intelligence Service and the Defense Science and Technology Agency have partnered with IBM to integrate quantum computing for complex mission planning, starting with logistics, such as military resupply missions. The goal is to determine whether quantum computing can eventually outperform conventional planning methods and prepare to harness quantum technologies for other uses as they mature.
7. Sustainably Managing Pests
Researchers at the University of Canterbury in New Zealand are using AI-powered sensors to test how natural methods can support sustainable pest control. Solar-powered sensors listen for and record certain sounds, and AI algorithms analyze the recordings to determine which species of birds and bats are in the area. Farmers then receive recommendations to attract beneficial birds and bats so they can reduce pesticide use.
8. Analyzing Brain Scans
Researchers at Boston University have created an AI tool that can recognize patterns in MRI brain scans. The tool can identify the different types of MRIs used to detect different diseases, clean up blurry images, erase patients’ skulls to provide an unobstructed view of the brain, and identify the different regions of the brain, speeding up the time it takes to perform and analyze an MRI scan.
9. Planning Fitness Routines
Exercise-equipment company Peloton is adding an AI chatbot to its app that will allow members to create or update their personalized fitness plan using plain language and text. The chatbot will ask users to describe their fitness goals in their own words and then propose a workout plan, including explanations of its logic. Users can make edits by continuing to converse with the chatbot.
10. Discovering Clean Energy Materials
Researchers at Tohoku University in Japan have developed a system that combines multiple AI technologies to identify promising clean energy materials faster and at a lower cost than conventional methods. The system continuously learns from the results of new experiments to improve its predictions and recommend the most promising candidates for further testing.
Image credit: Salam2009/Wikimedia
