Teen's AI Unearths 1.5 Million Cosmic Objects, Rewriting the Sky Map

High school student **Matteo Matthew Paz** has made a significant astronomical discovery, identifying approximately **1.5 million previously unrecognized objects** in space. Utilizing **artificial intelligence** and analyzing data from **NASA's NEOWISE mission**, Paz developed a system called **VARnet** that detects subtle, time-based infrared signals. This breakthrough, published in **The Astronomical Journal** and recognized with the top prize at the **Regeneron Science Talent Search**, challenges the notion that the sky has been fully mapped and opens new avenues for understanding cosmic phenomena like quasars and exploding stars. The sheer volume of data from NEOWISE, totaling **200 terabytes**, made manual analysis impossible, highlighting the power of AI in scientific exploration.

• A high school student used AI to identify 1.5 million new celestial objects.
• The discovery was made by analyzing data from NASA's NEOWISE mission.
• AI system VARnet detected previously missed time-based infrared signals.
• The findings challenge assumptions about the completeness of sky mapping.
• The breakthrough earned the student top honors and significant prize money.

The analysis of **NASA's NEOWISE** infrared data by **Matteo Matthew Paz** has resulted in the identification of approximately **1.5 million variable objects** that were not systematically cataloged. His AI system, **VARnet**, processed **200 terabytes** of data, achieving a high F1 score of 0.91 in identifying these celestial bodies. The study, published in **The Astronomical Journal**, highlights the challenge of analyzing large astronomical datasets and the efficacy of AI-driven pattern recognition in overcoming such obstacles. The significance lies in the systematic cataloging of these objects, rather than their sudden appearance.

This discovery is a monumental leap forward, demonstrating that even vast, seemingly insurmountable datasets can yield incredible insights with the right tools. **Matteo Matthew Paz's** success with **VARnet** validates the potential of AI in accelerating scientific discovery, promising a future where previously hidden cosmic wonders are routinely uncovered. The identification of **1.5 million new objects** will undoubtedly fuel new research into the dynamic nature of the universe, potentially leading to a more complete understanding of celestial evolution and phenomena.

While the discovery of **1.5 million objects** is impressive, it underscores a potential over-reliance on AI without sufficient human oversight. The fact that these objects were 'hidden in plain sight' within **NASA's NEOWISE** data for so long raises questions about the thoroughness of previous scientific methodologies and data analysis. Furthermore, the sheer scale of the identified objects could lead to an overwhelming influx of new research targets, potentially diluting focus and resources, and raising concerns about how these newly cataloged phenomena will be prioritized and studied effectively.

Originally reported by USA Herald