From High School to the Stars: Teen Uncovers 1.5 Million Cosmic Mysteries with AI
April 16th 2025
In a breakthrough that bridges youth, AI, and space science, Pasadena high school senior Matteo Paz stunned the astronomical community by using a machine-learning model he developed to uncover 1.5 million previously unknown celestial objects. Working at Caltech’s Summer Research Connection and mentored by astrophysicist Davy Kirkpatrick, Paz applied cutting-edge AI to NASA’s NEOWISE data—transforming a staggering archive of 200 billion datapoints into a goldmine of discovery.
Paz’s model was designed to detect variable cosmic objects, such as quasars and eclipsing binaries, often missed by traditional analysis. The AI not only streamlined this discovery process but created a reproducible tool other scientists could apply across astronomy and beyond—including environmental monitoring and even financial markets.
While the praise is widespread—and deserved—this achievement also raises deeper questions about the evolving role of AI in research. Does this mark a new era where youthful coders and smart algorithms rival traditional research teams? Or could it risk overdependence on machine filtering, potentially overlooking rare or complex phenomena that require human nuance to identify?
Supporters highlight the democratizing potential: AI offers young minds, regardless of institutional power or funding, the ability to contribute meaningfully to science. Yet some critics urge caution, noting that AI models trained on narrow parameters might filter out anomalies or interpret noise as signal—biases that require rigorous review to avoid false leads in scientific literature.
In the end, Paz’s story is not just about scientific stardom but about mentorship, curiosity, and the growing power of AI to amplify human potential. As he now works at Caltech and plans to publish a full catalog of discoveries, his story challenges us to rethink what age, expertise, and technology can mean in modern science.
Source: SciTechDaily
Matteo Paz with Caltech President Thomas F. Rosenbaum at the Regeneron Science Talent Search award ceremony on March 11, 2025. Rosenbaum, the Sonja and William Davidow Presidential Chair and professor of physics, gave an address at the event as the Society for Science board chair and a Science Talent Search alumnus. Credit: Society for Science / SciTechDaily