Feed: Slashdot
Feed Link: https://slashdot.org/
---

Title: America's Leading Alien Hunters Depend on AI to Speed Their Search

Link: https://science.slashdot.org/story/25/05/23/1...

Harvard University's Galileo Project is using AI to automate the search for
unidentified anomalous phenomena, marking a significant shift in how
academics approach what was once considered fringe research. The project
operates a Massachusetts observatory equipped with infrared cameras, acoustic
sensors, and radio-frequency analyzers that continuously scan the sky for
unusual objects. Researchers Laura Domine and Richard Cloete are training
machine learning algorithms to recognize all normal aerial phenomena --
planes, birds, drones, weather balloons -- so the system can flag genuine
anomalies for human analysis. The team uses computer vision software called
YOLO (You Only Look Once) and has generated hundreds of thousands of
synthetic images to train their models, though the software currently
identifies only 36% of aircraft captured by infrared cameras. The Pentagon is
pursuing parallel efforts through its All-domain Anomaly Resolution Office,
which has examined over 1,800 UAP reports and identified 50 to 60 cases as
"true anomalies" that government scientists cannot explain. AARO has
developed its own sensor suite called Gremlin, using similar technology to
Harvard's observatory. Both programs represent the growing legitimization of
UAP research following 2017 Defense Department disclosures about military
encounters with unexplained aerial phenomena.

Read more of this story at Slashdot.

---
VRSS v2.1.180528