An artificial intelligence (AI) model has been used to create a new image of the black hole at the center of the Milky Way galaxy, with some experts expressing concern over its accuracy.

According to the AI model, the supermassive black hole at the heart of our galaxy is spinning at an incredibly fast rate, based on data that was previously considered too noisy to be useful. The goal of this model is to produce the most detailed images of black holes to date, but not all experts are convinced of its reliability due to the questionable quality of the data.

Reinhard Genzel, an astrophysicist at the Max Planck Institute for Extraterrestrial Physics in Germany and a 2020 Nobel Prize winner in physics, expressed interest in the AI model but cautioned that artificial intelligence is not a cure-all solution.

Scientists have long sought to study Sagittarius A*, the black hole at the center of our galaxy. In 2022, the first image of this massive object was revealed, but questions remained about its behavior.

An international team of scientists has now turned to AI to extract more information about Sagittarius A* from data collected by the Event Horizon Telescope (EHT). Unlike traditional telescopes, the EHT is a network of instruments around the world that work together to observe black holes using long electromagnetic waves.

However, the EHT's technique is sensitive to interference from Earth's atmosphere, making it challenging for researchers to interpret the data. Michael Janssen, an astrophysicist at Radboud University and a study co-author, believes that a neural network is well-suited to address this issue.

Janssen and his team trained an AI model on noisy EHT data that had previously been discarded. By doing so, they were able to generate a new image of Sagittarius A* that revealed previously unseen features, such as the black hole's high rotational speed and its axis pointing towards Earth.

Despite these findings, some experts, including Genzel, remain skeptical of the AI model's accuracy due to the low-quality data used. They suggest that the new image may be distorted and caution against accepting it without further validation.

In the future, Janssen and his team plan to refine their AI technique using the latest EHT data and compare it to real-world observations to enhance the model's accuracy and reliability.