Colour photographs now represent almost all the images produced with the new reality capture tools, mobile phones, which in 2020 ‘took’ 90% of all photos of that year. Black and white is relegated to artistic expression, even newspapers have converted to colour for some years. In the history of photography, although research on colour is attempted from the early stages, it is necessary to wait until 1861 with the experiences of James Clerk Maxwell who created a stable colour image. However, it is from the fifties of the twentieth century that the use of colour becomes ‘popular’ even in a more aesthetic dimension than an objective reproduction of reality. Part of the ethnographic, anthropological, archaeological and field research, on the other hand still makes use of consolidated and inexpensive black and white for a long time. On these images largely available online and open source you can conduct automatic colouring experiences. The procedure, managed with artificial intelligence algorithms with deep learning processes, is always more widely used with free applications and allows to obtain qualitatively more and more relevant results, even if some critical analysis is still necessary. This article presents the state of the art to 2021 of automatic colouring, with the comparison between algorithms developed since 2016 and showing with experimental examples both the possibilities of rendering and even the critical issues that emerged with the application in anthropological photographs, with the aim of extracting information that is not very evident in the originals in black and white.