Besides the "lightness" of the article (which could be expected as arstechnica is not a publication aimed to construction engineers) the actual paper (PDF) can

Besides the "lightness" of the article (which could be expected as arstechnica is not a publication aimed to construction engineers) the actual paper (PDF) can be downloaded from Nature: https://www.nature.com/articles/s41598-023-32981-y.pdf https://www.nature.com/articles/s41598-023-32981-y.pdf is perplexing. Of course (omitted in the arstechnica article) the used diapers, besides being washed and dried are shredded. Then, using 370 kg/m3 of cement (which is on the high side for 25 MPa concrete) they fail to obtain 25 MPa, but having obtained 24.91 Mpa (on a meaningless amount of samples) they call it a success. Then they go for a tangent, applying (probably legal in Indonesia) laughingly low compression strength requirements to structural components such as columns and beams without any considerations about the interaction with reinforcement, permeability, cracking behaviour.