Researchers have created an AI-powered video of a black hole jet using over 100 images taken of a distant cosmic object known as a blazar. The video, compiled from 116 images captured between 1995 and 2022, shows the jet of the blazar 3C 345, located in the constellation Hercules. A blazar is a special type of quasar — a galaxy with a supermassive black hole at its center that is actively feeding — which emits powerful jets of gas at nearly the speed of light. These jets are filled with high-energy radiation, including X-rays and gamma rays, and are of great interest to astronomers. The study, published in the journal Nature on August 26, used data from the Very Long Baseline Array (VLBA), a network of 10 radio telescopes spread across the United States. The VLBA has been observing blazar jets for decades through two programs, BEAM-ME and MOJAVE, which have tracked hundreds of such sources. To transform these images into a high-resolution video, the researchers used an AI neural network called Kine. This model improved the resolution of the video four times beyond that of any single image, allowing scientists to measure the jet’s speed with unprecedented precision. The results were surprising. The brightest parts of the jet were moving at 10 to 13 times the speed of light, while the surrounding gas was moving at about nine to 12 times the speed of light. This contradicts the general expectation that the bright components — thought to be shock waves — should move faster than the surrounding material. While the study does not rule out the shock model entirely, it raises questions about its application to this specific source. The VLBA provides a broad view of the universe but lacks the fine detail needed for such precise measurements. The AI model Kine helps bridge this gap by creating high-resolution videos that capture changes in brightness over time. As a neural network, Kine mimics the human brain by learning from data through layers of computational nodes. It can analyze observations at different times, identifying patterns in both space and time. The researchers are excited about the potential of Kine for future studies. They believe it could transform how scientists study jet dynamics in space. By precisely measuring the velocity of the jet at any point, the method offers a new tool for understanding the behavior of black hole jets and their impact on their surroundings.