7AdvancedExamScience

Teen Scientist's Algorithm Reveals Millions of Hidden Cosmic Objects

An eighteen-year-old high school student from California developed an artificial intelligence algorithm that identified 1.5 million previously unknown cosmic objects in archived NASA data. His peer-reviewed research, published as a single-author paper in The Astronomical Journal, demonstrates how advanced machine learning can transform existing datasets and reshape modern astrophysics.

2 min readLevel 7March 23, 2026

By Flalingo Editorial Team

Vocabulary

Study these words before reading the article. Click the audio button to hear pronunciation.

paradigm

/ˈpærəˌdaɪm/

noun

A typical example, pattern, or model of something; a fundamental framework of thinking or practice in a particular field.

“The discovery of AI-driven research methods represents a paradigm shift in modern scientific inquiry.”

untapped

/ʌnˈtæpt/

adjective

Not yet used, exploited, or taken advantage of; available but not yet utilised.

“The vast archive of infrared data had remained untapped for over a decade before Paz analysed it.”

decomposition

/ˌdiːkɒmpəˈzɪʃən/

noun

The process of breaking something down into smaller or simpler components, often for analysis.

“Wavelet decomposition allowed the algorithm to separate genuine signals from instrument noise.”

substantive

/ˈsʌbstəntɪv/

adjective

Having real importance, value, or significance; meaningful rather than superficial.

“Her research made a substantive contribution to the understanding of climate change patterns.”

facilitate

/fəˈsɪlɪteɪt/

verb

To make a process or action easier or more achievable; to help bring about a result.

“The publicly released catalog is expected to facilitate new research into stellar evolution.”

verification

/ˌverɪfɪˈkeɪʃən/

noun

The process of confirming the truth, accuracy, or validity of something through evidence or investigation.

“Each candidate object requires independent verification before it can be classified as a confirmed discovery.”

Article

Click on any underlined word to see its definition. Highlighted words are from the vocabulary section.

Seldom has a scientific contribution from a teenager generated such profound implications for contemporary astrophysics. Matteo Paz, an eighteen-year-old student from Pasadena, California, developed an AI algorithm that uncovered approximately 1.5 million previously unidentified cosmic objects. His model processed roughly 200 billion infrared detections gathered over a decade by NASA's now-retired NEOWISE telescope. The findings were published in The Astronomical Journal in November 2024, with Paz listed as the sole author.

The NEOWISE telescope was originally launched in 2009 to detect near-Earth asteroids and comets. However, it simultaneously collected vast infrared data on distant variable objects, including quasars, exploding stars, and eclipsing binary systems. Much of this supplementary data had remained untapped for years. Paz's mentor, Caltech senior scientist Davy Kirkpatrick, had initially envisioned a modest manual analysis of a small portion of the sky.

Rather than pursuing a conventional approach, Paz constructed a machine-learning pipeline called VARnet within just six weeks. The architecture integrates wavelet decomposition, Fourier transforms, and a convolutional neural network to classify variable sources. This sophisticated methodology enabled the algorithm to detect subtle brightness fluctuations that traditional tools had overlooked. Each flagged source constitutes a candidate requiring subsequent observational verification by professional astronomers.

Paz's achievement earned him the first-place prize of 250,000 dollars in the 2025 Regeneron Science Talent Search. The resulting catalog, released publicly in late 2025, is already facilitating observations at major facilities such as the Vera C. Rubin Observatory. Furthermore, NASA leadership has formally acknowledged the significance of his AI-driven breakthrough. Paz is now employed at Caltech's Infrared Processing and Analysis Center while completing his secondary education.

What distinguishes this accomplishment is its broader implications for scientific methodology and accessibility. The algorithm's adaptability suggests potential applications in fields such as financial analysis, environmental monitoring, and neuroscience. Paz developed his advanced mathematical skills through a public programme for gifted students in Pasadena. His trajectory underscores a paradigm shift in which democratised artificial intelligence empowers individuals outside institutional frameworks to make substantive contributions.

Comprehension Quiz

Test your understanding of the article with these multiple-choice questions.

1
What was the primary achievement described in the article?
2
What was the original mission of the NEOWISE telescope?
3
How long did it take Paz to build the initial machine-learning pipeline?
4
Which prize did Paz win for his research?
5
According to the article, what potential applications does Paz's algorithm have beyond astronomy?

Discussion

Answer these comprehension questions about the article.

1
According to the article, what was the original purpose of the NEOWISE telescope, and how did Paz's research extend its scientific value?
2
The author suggests that Paz's methodology was more ambitious than what his mentor initially proposed. What evidence from the passage supports this claim?
3
What are the three integrated stages of VARnet's architecture, and why was each stage necessary for processing the NEOWISE data?
4
According to the passage, why does the figure of 1.5 million objects not represent confirmed discoveries in the traditional sense?
5
How does the article characterise the broader significance of Paz's achievement beyond the field of astronomy?

Further Discussion

Share your opinions on these open-ended questions.

1
To what extent do you believe artificial intelligence will replace traditional methods of scientific research in the coming decades? Justify your position with examples.
2
The article highlights the role of mentorship in Paz's success. How important has mentorship or guidance from others been in your own educational or professional development?
3
Should publicly funded scientific data, such as that collected by NASA, be made more accessible to citizen scientists and students? What are the potential benefits and risks?
4
Paz developed graduate-level skills while still in secondary school. Do you think the current education system adequately supports exceptionally gifted students? Why or why not?
5
If you had access to a powerful AI algorithm capable of analysing large datasets, which real-world problem would you choose to investigate, and why?