On 19 March 2025, the European Space Agency's (ESA) Euclid mission released the first batch of data from its mapping, including a preview of its deep fields. Here, hundreds of thousands of galaxies of different shapes and sizes take centre stage and offer a glimpse of their large-scale organisation in the cosmic web. These galaxies, observed and analysed by scientists from the Euclid Consortium, demonstrate the unprecedented power of this telescope. The scientific results are described in a series of 27 scientific publications, along with 7 technical papers describing the processing of these data, published today in arXiv.

‘Euclid proves once again to be the ultimate discovery machine. It is studying galaxies on large scales, allowing us to explore our cosmic history and the unseen forces shaping our Universe,’ says Carole Mundell, ESA's Director of Science. ‘With the release of the first Euclid mapping data, we are unveiling a treasure trove of information that scientists will be able to analyse and address some of the most intriguing questions in modern science. In doing so, ESA is fulfilling its commitment to drive scientific progress for future generations.

‘The data we are publishing now represent a small fraction of the total data that Euclid will collect,’ says Francisco Javier Castander, researcher at the Institute of Space Sciences (ICE-CSIC) and the Institute of Space Studies of Catalonia (IEEC). ‘However, even such a small fraction of the data has allowed us to carry out many scientifically relevant studies, which we now present.

Euclid is expected to capture images of more than 1.5 billion galaxies over six years, sending around 100 GB of data daily. Such an impressively large dataset creates incredible opportunities for discovery, but also enormous challenges in finding, analysing and cataloguing galaxies. The advancement of artificial intelligence (AI) algorithms, in combination with thousands of volunteers and citizen science experts, is playing a key role.

In this sense, the study led by Marc Huertas-Company, researcher at the Instituto de Astrofísica de Canarias (IAC), used Euclid's unique combination of wide field of view and resolution, together with the models trained thanks to volunteers from all over the world who participated in the citizen science campaign, to quantify the abundance of bars and their evolution with cosmic time. A study in which CEFCA researcher Jesús Vega has collaborated.

The publication of the Q1 data demonstrates the transformative power of the Euclid telescope for the study of galactic physics. It also confirms the growing impact of AI in the analysis of increasingly voluminous and complex data in astrophysics. Approximately 50% of the scientific papers published with these early Euclid data are based on AI methods,’ adds Marc Huertas-Company.

The other study in which Jesús Vega has collaborated, led by IAC researcher Malgorzata Siudek, presents AstroPT, a powerful AI model that learns from images and light patterns (SED) of galaxies, without the need for large amounts of human-labelled data. Thanks to Euclid's ultra-sharp images and deep infrared observations of the sky, the AI model can learn from higher quality and more diverse data than ever before.

‘We explore the use of foundational AI models, similar to those used in Chat GPT, for data exploration. These intelligent models are able to perform a multitude of tasks for which they were not specifically trained and open a new avenue for the use of AI in science,’ says Malgorzata Siudek. Euclid observations, due to their volume and complexity, are ideal for the development of this new generation of AI.