An astronomy professor at the University of Arizona wrote that artificial intelligence (AI) has begun to help astronomers tame large datasets and uncover their secrets.
The clarity of the famous first image of a black hole has been improved by a factor of two. A research team used AI to greatly enhance the first image from 2019, now showing a darker and larger black hole at the center of the M87 galaxy.
"I am an astronomer who has researched and written articles on cosmology, black holes, and exoplanets. Astronomers have been using AI for decades. In fact, in 1990, I was one of the first astronomers at the University of Arizona, where I served as a professor, to use a form of AI called neural networks to study the shapes of galaxies."
Since then, AI has permeated various fields of astronomy. As this technology becomes more powerful, AI algorithms have begun to help astronomers tame large datasets and discover new knowledge about the universe.
Better telescopes, more data
Since the inception of astronomy as a science, it has been attempting to understand the multitude of objects in the night sky. When the only tools available were the naked eye or a simple telescope, it was relatively simple, and only a few thousand stars and a handful of planets could be seen.
A hundred years ago, Edwin Hubble used newly constructed telescopes to demonstrate that the universe is not only filled with stars and gas clouds but also countless galaxies. As telescopes continue to improve, the number of celestial bodies visible to humans and the amount of data that astronomers need to organize have grown exponentially.
For example, the upcoming Vera C. Rubin Observatory in Chile will capture images so large that it would take 1,500 high-definition TV screens to fully view each image. It is expected to generate 0.5 EB of data within 10 years, approximately 50,000 times the information contained in all the books in the Library of Congress.
There are 20 telescopes with mirror diameters greater than 20 feet (6 meters). AI algorithms are the only hope for astronomers to handle all available data. AI has many useful methods for processing this data.
Picking out patterns
Astronomy often involves searching for a needle in a haystack. Approximately 99% of pixels in astronomical images contain background radiation, light from other sources, or the darkness of space, while only 1% exhibit the subtle shapes of faint galaxies.
AI algorithms, especially neural networks with interconnected nodes capable of learning to recognize patterns, are well-suited for picking out patterns of galaxies. Astronomers began using neural networks to classify galaxies in the early 2010s. These algorithms are now highly effective, achieving a 98% accuracy rate in classifying galaxies.
This story has also been repeated in other areas of astronomy. Astronomers involved in the Search for Extraterrestrial Intelligence (SETI) use radio telescopes to search for signals from distant civilizations. In the early days, radio astronomers scanned charts by eye, looking for unexplained anomalies. More recently, researchers have utilized 1.5 million personal computers and 1.8 million citizen scientists to search for artificial radio signals. Now, researchers are using AI to sift through vast amounts of data faster and more thoroughly than humans, allowing SETI efforts to cover more ground while greatly reducing the number of false positive signals.
Another example is the search for exoplanets. Astronomers have discovered the majority of the 5,300 known exoplanets by measuring the decrease in starlight as a planet passes in front of its star. AI tools can now identify signs of exoplanets with 96% accuracy.
Creating new discoveries
AI has proven itself to be excellent at identifying known objects in astronomy, such as galaxies or exoplanets. However, it is also powerful in searching for objects or phenomena that have not yet been discovered in the real world.
Research teams have used this approach to detect new exoplanets, understand the ancestral stars that led to the formation and growth of the Milky Way, and predict the characteristics of new types of gravitational waves.
To do this, astronomers first use AI to convert theoretical models into observable features, including realistic levels of noise. Then, they use machine learning to improve the AI's ability to detect and predict phenomena.
Finally, radio astronomers have also been using AI algorithms to screen for signals that do not match known phenomena. Recently, a research team from South Africa discovered a unique object that may be the remnants of two supermassive black holes merging. If confirmed, this data will allow for new tests of Albert Einstein's theory of general relativity, which describes the nature of spacetime.
Making predictions and filling gaps
Just as in many other areas of life, generative AI and large language models like ChatGPT have also made waves in the field of astronomy.
The team that created the first image of a black hole in 2019 used generative AI to produce new images. To do this, they first trained the AI to recognize black holes by providing it with various simulations. Then, the research team used their AI model to fill in the gaps in the vast amount of data collected by radio telescopes on the black hole M87.
Using these simulated data, the team was able to create a new image that is twice as clear as the original, and it perfectly matches the predictions of general relativity.
Astronomers are also turning to AI to help tackle the complexity of modern research. A team at the Harvard-Smithsonian Center for Astrophysics created a language model called astroBERT for reading and organizing 15 million scientific papers in the field of astronomy. Another team at NASA has even proposed using AI to prioritize astronomical projects, a process that astronomers undertake only once every 10 years.
As AI continues to develop, it has become an essential tool for astronomers. With telescopes improving, datasets growing, and AI constantly advancing, this technology is likely to play a central role in future cosmic discoveries.
Alphabet Q1 paid subscribers reached 350M, fueled by YouTube & Google One. YouTube ad revenue slightly missed at $9.88B amid Premium growth. Total Q1 revenue hit $109.9B, beating forecasts.
Anthropic releases Claude connectors for Adobe, Blender, and Ableton, enabling AI workflow automation and app control. It also sponsors the Blender Development Foundation.
Apple introduces an App Store model splitting 12-month plans into monthly installments. On iOS 26.4+, it lowers upfront costs without reducing total price, adding billing tracking.
Google launches Aadhaar digital IDs in Wallet for India, enabling secure QR/NFC verification. A phased partner rollout targets key sectors, while passport-based credentials expand globally.
YouTube tests "Ask YouTube," an AI tool blending text and video for complex queries. Initially for US Premium users, it supports interactive follow-ups, expanding Google's AI search.
Google grants US DoD classified AI access, joining OpenAI and xAI in defense deals amid Anthropic's dispute. Contractual safeguards against surveillance and lethal weapons remain unenforceable.
Mistral AI launches Workflows, a Temporal-powered orchestration platform for secure, Python-driven enterprise AI automation featuring human-in-the-loop approvals.