Carnegie Mellon Launches AI-Astronomy Fellowship to Accelerate Cosmic Discovery

Carnegie Mellon University has launched the Keystone Astronomy & AI (KAAI) Visiting Fellows Program, a new initiative supported by the Simons Foundation that aims to accelerate the application of artificial intelligence to cosmological and astronomical research.

How the Program Works

KAAI Fellows participate in a monthlong residency at the McWilliams Center for Cosmology & Astrophysics, where each visiting fellow is paired with two mentors — one in astrophysics and one in AI or statistics. The dual-mentorship structure is designed to ensure that research projects draw meaningfully from both disciplines rather than simply applying off-the-shelf machine learning tools to astronomical data.

Each residency culminates in a hands-on workshop that shares software, datasets, and workflows with the broader research community. The initiative also provides opportunities for Carnegie Mellon graduate students, who collaborate with visiting fellows and gain direct experience applying AI to frontier problems in astrophysics.

Why AI and Astronomy

"AI is changing how we do science, and astronomy is where its impact will be felt first and fastest," said Tiziana Di Matteo, director of the McWilliams Center. The statement reflects a growing recognition that the scale of astronomical data — from next-generation telescopes, gravitational wave detectors, and space-based observatories — has exceeded what traditional analysis methods can handle.

Machine learning techniques are already being applied to tasks like galaxy classification, exoplanet detection, and gravitational lens identification. But the KAAI program aims to push beyond these established applications toward more fundamental challenges: using AI to build better physical models of the universe, extract signals from noisy data, and identify patterns that human researchers might miss.

A Model for Cross-Disciplinary AI Research

The program represents a template for how AI research can be structured at the intersection of machine learning and domain science. Rather than building a permanent research group, the visiting fellows model brings diverse perspectives and expertise into a concentrated collaborative environment, producing shared tools and published research that benefit the wider community.

With the KAAI program, Carnegie Mellon is positioning itself at the center of a growing movement to apply AI not just to engineering and commercial problems, but to fundamental scientific questions about the nature of the universe.