The first image of a black hole, which captivated the world in 2019, has received a remarkable upgrade thanks to the application of artificial intelligence (AI). A research team utilized AI algorithms to enhance the initial image, resulting in a sharper depiction that showcases the black hole at the center of the M87 galaxy as darker and larger than before.

AI has been an integral part of astronomy for decades, with astronomers harnessing its power to study galaxies and explore cosmology. In 1990, researchers from the University of Arizona pioneered the use of neural networks, a form of AI, to analyze galaxy shapes. Since then, AI has permeated every aspect of astronomy, helping scientists navigate vast data sets and uncover new insights about the universe.

Throughout the history of astronomy, understanding the multitude of celestial objects has been a complex endeavor. The advancements in telescopes have led to a tremendous increase in the number of observable celestial entities, resulting in exponential growth in data volume. For instance, the forthcoming Vera Rubin Observatory in Chile is expected to generate an astonishing 0.5 exabytes of data over a decade, equivalent to around 50,000 times the information stored in the Library of Congress.

With AI algorithms, astronomers can tackle the challenges of processing this massive amount of data. AI, particularly neural networks with their ability to recognize patterns, excels at identifying faint galaxies amidst the vast background radiation and other sources of light. The accuracy of these algorithms has reached an impressive 98% in classifying galaxies.

The impact of AI extends beyond galaxy classification. In the Search for Extraterrestrial Intelligence (SETI), AI has facilitated the swift and thorough analysis of radio signals, enabling broader coverage and a significant reduction in false positive signals. The search for exoplanets has also benefited from AI’s prowess, as it can identify the telltale signs of exoplanets with 96% accuracy.

Furthermore, AI has proven invaluable in the realm of theoretical astronomy. By converting theoretical models into observational signatures using AI, astronomers can detect new exoplanets, unravel the origins of the Milky Way, and predict the characteristics of gravitational waves. AI has even uncovered unique phenomena, such as a potential remnant of the merging of two supermassive black holes, offering an opportunity to test Albert Einstein’s general relativity.

Generative AI and large language models, including ChatGPT, have also made their mark in astronomy. The team behind the groundbreaking black hole image used generative AI to produce a new, sharper image. By training an AI model on simulations of various black holes, they utilized the model to fill in the gaps in the extensive data collected by radio telescopes. The result was an image twice as sharp as the original, aligning perfectly with the predictions of general relativity.

Moreover, AI has become a crucial tool for handling the complexity of modern research. Language models like astroBERT have been developed to read and organize vast scientific papers, aiding in the organization and accessibility of astronomical knowledge. NASA has even proposed using AI to prioritize astronomy projects, streamlining the decision-making process.

As AI continues to advance alongside the improvement of telescopes and the expansion of data sets, it is clear that this technology will play a central role in uncovering the mysteries of the universe. With sharper images, accelerated data processing, and enhanced research capabilities, AI empowers astronomers to push the boundaries of our understanding and embark on new discoveries.

By Impact Lab